Abstract

Research relating to minority stressors generally explores mental health outcomes, with limited focus on the physical dimension. In addition, minority stress research is conducted mainly in Christian-oriented societies. To address these pitfalls we used Web sampling targeting Israeli participants ages 12 to 30 (N = 952; 28 percent heterosexuals, 78 percent lesbian, gay, and bisexual [LGB] adolescents and young adults) to assess their mental health, physical and sexual risk behaviors, minority stressors, and coping resources. Results indicate that young LGBs had lower levels of mental and physical health than heterosexuals. Among LGB participants, high levels of minority stressors and low levels of coping resources predicted lower levels of mental health, and lower levels of mental health predicted lower levels of physical health. These results emphasize that minority stressors should be recognized as risk factors for poorer mental health, as well as for physical and sexual risk behaviors.

Studies have shown significant health disparities between lesbian, gay, and bisexual (LGB) individuals and their heterosexual counterparts—including higher incidence of substance use, mental and physical health problems, and greater involvement in sexual risk behaviors (Cochran, Sullivan, & Mays, 2003; Horvath, Remafedi, & Rosser, 2008). Specific attention has been given in research to LGB youths and young adults (comprising adolescents and young adults ages up to 30)—a population known to be at risk for mental health disparities (D’Augelli, 2006). These disparities may be related to the destructive effects of the stigma, victimization, and isolation that sexual minorities endure in heteronormative, occasionally homophobic society, and to the psychological stressors involved in acknowledging and accepting their sexual orientation (Rivers & D’Augelli, 2001).

Minority stress theory (Meyer, 2003) maintains that the heightened vulnerability of LGB individuals stems from their subjection to unique stressors resulting from their minority status. These include distal objective stressors—reflecting the level of heterosexism in the environment (such as antigay harassment)—and subjective proximal stressors (such as internalized homophobia and concealing one’s sexual orientation), reflecting the LGB individual’s perception of the environment as threatening. The impact of such minority stressors can be alleviated by certain coping resources—especially social support and connectedness to the LGB community (Kertzner, Meyer, Frost, & Stirratt, 2009; Meyer, 2003). Risky sexual behaviors, which include anal/vaginal sex without a condom with a partner whose HIV status is discordant or unknown, and sexual intercourse under the influence of alcohol or other substances, may expose individuals to HIV and other sexually transmitted diseases (Mor, Davidovich, McFarlane, & Feldstein, 2008). Physical risk behaviors such as smoking and substance use, just as risky as these sexual behaviors (Harte & Meston, 2012), are often explained by the individual’s mental status in response to subjective and objective stressors due to their disadvantaged social status, and suggest a link between mental and physical health (Hass, Schefer, & Kornienko, 2010; Meyer, 2003). Minority stress approach is now contested by researchers, who argue that these deficit models of LGB development infer illness, and call for further investigation of resilience factors (such as family and friends’ support) effecting positive aspects of mental health among young LGBs (Savin-Williams, 2005). As suggested by Shilo & Savaya (2011), a resilience–risk continuum may explain the impacts of social stressors and coping resources on the health of LGB youths and young adults. Whereas resiliency is described as the set of protective factors that make a person resistant to the detrimental effects of stressors, vulnerability represents the opposite end of the continuum (Ingram & Price, 2001), and both are fundamental to the understanding of minority stress approach.

Nationally conducted random surveys (Russell & Joyner, 2001) as well as nonrandom studies worldwide (D’Augelli, 2006; Horvath et al., 2008) have linked adverse mental health outcomes among LGB youths and young adults, including suicidal ideation and behavior, to stressors such as violence, bullying, and harassment due to their sexual orientation. Evidence also suggests that this population—particularly adolescent and young adult gay men—is at high risk for substance use and sexual risk taking (Wong, Kipke, Weiss, & McDavitt, 2010). Research relating to minority stress theory generally looks at mental health outcomes with limited focus on the physical dimension (Kertzner et al., 2009; Meyer, 2003).

Recent studies have shown that LGB youths and young adults in Israel, a developed country characterized by a relatively openly gay society, disclose their sexual orientation at the mean age of 16.5 to both friends and family members (Shilo & Savaya, 2011) and that 67.5 percent of gay men had their first sexual encounter before 18 years of age (Mor, Davidovich, Bessudu-Manor, & McFarlane, 2012). It is also known that the annual number of newly diagnosed HIV infections among the LGB population in Israel has tripled between 1999 and 2010 (Mor & Dan, 2012). With the growing concern over the increase of sexual and risk behaviors among LGB individuals in Israel, and given the limited data available on the physical health behavior among sexual minority youths, the present study set out to compare the mental and physical health (comprising physical and sexual risk components) of Israeli LGB youths and young adults with that of their heterosexual counterparts, and define the predictors of mental and physical health among young LGB youths and young adults. We hypothesized that (1) young LGB would experience poorer mental and physical health than heterosexuals; (2) demographic characteristics (such as being an adolescent or male) and stressors related to belonging to a sexual minority (internalized homophobia, level of outness, antigay harassment) would be associated with poorer mental and physical health outcomes, while coping resources (support from family and friends, and connectedness to the LGB community) would be associated with more favorable mental and physical health outcomes; and (3) poorer mental health would be associated with risky behaviors.

Method

Procedures

Study participants were recruited between July and October 2010, and were asked to complete a Hebrew-language electronic questionnaire hosted on a secure Web page (https://send2.bignet.co.il/expo4u/). Due to the difficulty of sampling LGB individuals, with the lack of sampling frame, an online venue sampling (Meyer & Wilson, 2009) was used by offering the questionnaire to members of nine LGB groups on Facebook (among which six were for LGB youths and young adults, two for young gay males, and one for young lesbians) and six other young LGB Web forums. Similarly, heterosexual participants were recruited by offering the questionnaire to members of seven general groups on Facebook and six general Web forums aimed at young participants. To reduce potential bias, Facebook groups and Web forums for LGB youths and young adults and young heterosexuals were matched on the basis of their content (for example, “LGBs for human rights” and “young Israelis for human rights” Facebook groups; “young LGBs” and “up to 30” Web forums, both for social contact among forum participants). In addition, Web forums and Facebook groups were excluded from our sampling frame if they were likely to overrepresent people seeking or receiving support for mental or physical health problems, or people seeking sexual encounters (for example, a group that offers help for people with HIV, a Web forum that offers help and support for young LGBs during the “coming out” process, and LGB and heterosexual dating forums). Eligibility criteria were being a Jewish adolescent or young adult (ages 12 to 30) living in Israel. The purpose of the study was stated on page 1 of the questionnaire: to study the various aspects of growing up in Israel as a young heterosexual, gay, lesbian, or bisexual. Participants were asked to confirm consent electronically before completion. All study procedures were reviewed and approved by the institutional review boards of Tel Aviv University, the Wolfson Medical Center, and the Israel Gay Youth Organization.

A potential pitfall in Web-based sampling—that it may exclude segments of the population studied due to different computer accessibility and use (Meyer & Wilson, 2009)—is not really applicable to this study. Ninety-two percent of Israelis 30 years old or younger have access to the Internet and use it on a daily basis (Central Bureau of Statistics, 2009).

Participants

A total of 952 participants, comprising almost equal numbers of males (n = 508, 53.4 percent) and females (n = 444, 46.6 percent), completed the questionnaire, with a mean age of 22.1 ±4.7 years. Participants reported their age, gender, sexual orientation (gay or lesbian, bisexual, heterosexual, questioning) and place of residence. On the basis of these responses, participants were classified as either adolescent (age ≤ 18) or young adult (age 19 and older), LGB (including questioning) or heterosexual, and residing in greater Tel Aviv/central region or in the periphery. Age classification was based on the fact that in Israel (like in other Western countries) the age of 18 legally marks the transition from adolescence to adulthood. In Israel, it is also the age when most Jewish youths are conscripted into the compulsory military service, which is a significant psychological and social phase in the lives of young Jewish Israelis (Dar & Kimhi, 2001).

Of the total 952 participants, 685 (72 percent) self-identified as LGB and 267 (28 percent) as heterosexuals. Demographic characteristics of the LGB participants are presented in Table 1. Most participants (both heterosexual and LGB) described themselves as secular (n = 830, 86.2 percent), while 133 (13.8 percent) described themselves as “traditional” or “religious.” Collectively, participants resided in 104 cities and towns across Israel, representing all seven social demographic clusters in Israel (Central Bureau of Statistics, 2009). Of all study participants, 222 (23.3 percent) were high school students, 335 (35.2 percent) were attending university, 106 (11.1 percent) were in compulsory military service, and 523 (54.9 percent) were employed. Furthermore, 334 participants (35.1 percent) were both students (at high school or university) and employed. No statistical group differences were found in sociodemographic characteristics (age, religiosity, occupation, and residency) between heterosexual and LGB participants.

Table 1:

Demographic Characteristics among LGB Participants, by Sexual Orientation (N = 685)

 Gay Boys (n = 390)
 
Lesbians (n = 142)
 
Bisexuals (n = 153)
 
Variable n (%) M (SD) n (%) M (SD) n (%) M (SD) 
Age  22.9 (4.5)  22.4 (4.7)  21.5 (4.4) 
Gender       
 Male 390 (100)    51 (33.3)  
 Female   142 (100)  102 (66.7)  
Religiosity       
 Secular 348 (89.5)  119 (84.4)  137 (89.5)  
 Traditional 35 (8.7)  19 (13.5)  11 (7.2)  
 Religious 7 (1.8)  3 (2.1)  5 (3.3)  
Occupation       
 High school 73 (18.7)  41 (28.9)  32 (20.9)  
 Military service 46 (11.8)  18 (12.7)  13 (8.5)  
 University 145 (37.2)  48 (33.8)  34 (22.2)  
 Employed 234 (60) 74 (52.1) 86 (56.2) 
 Gay Boys (n = 390)
 
Lesbians (n = 142)
 
Bisexuals (n = 153)
 
Variable n (%) M (SD) n (%) M (SD) n (%) M (SD) 
Age  22.9 (4.5)  22.4 (4.7)  21.5 (4.4) 
Gender       
 Male 390 (100)    51 (33.3)  
 Female   142 (100)  102 (66.7)  
Religiosity       
 Secular 348 (89.5)  119 (84.4)  137 (89.5)  
 Traditional 35 (8.7)  19 (13.5)  11 (7.2)  
 Religious 7 (1.8)  3 (2.1)  5 (3.3)  
Occupation       
 High school 73 (18.7)  41 (28.9)  32 (20.9)  
 Military service 46 (11.8)  18 (12.7)  13 (8.5)  
 University 145 (37.2)  48 (33.8)  34 (22.2)  
 Employed 234 (60) 74 (52.1) 86 (56.2) 

Measures: Outcome Variables

Mental health was assessed by the Mental Health Inventory (MHI) (Veit & Ware, 1983), a 38-item measure of mental distress (25 items) and psychological well-being (13 items). Items were rated on a six-point scale ranging from 1 (strongly agree) to 6 (strongly disagree). The MHI was translated into Hebrew by Florian & Drori (1990), who confirmed both construct and external validity of the Hebrew version on the basis of a representative sample of Israeli population and reported high internal consistency reliability (α = .95, α = .93 for distress and well-being scales, respectively). The present study established a reliability of α = .91 for the distress scale and α = .94 for the well-being scale. Scores were calculated as the sum of the item measures of each index: The higher the score, the greater the distress and well-being.

Physical health was assessed by 11 measures of known physical health risk behavior and sexual risk behavior. Tobacco, alcohol, and other drug use are the leading causes for heart disease and cancer, which leads to high rates of mortality and morbidity in all Organisation for Economic Co-operation and Development countries; these behaviors are often established during youth and extended into adulthood (Kolbe et al., 2004). Therefore, physical health risk behavior was gauged in this study by means of eight questions, adapted from the Centers for Disease Control and Prevention’s Youth Risk Behavior Surveillance System (CDC-YRBSS) (Eaton, Kann, Kinchen, Shanklin, & Flint, 2012), pertaining to smoking (more than four cigarettes a day), using drugs without prescription, using anabolic steroids to gain muscle hypertrophy, or substance use (alcohol above one daily portion, club drugs, chemical stimulants, or hallucinogens) in the previous six months. Sexual risk behavior was assessed by three questions, based on the CDC-YRBSS, relating to the practice of anal intercourse without a condom and to the use of drugs or alcohol, respectively, before or during sexual intercourse within the previous six months. These behaviors were found to expose heterosexual males and females, as well as gay males and lesbians, to HIV and other sexually transmitted diseases (Everett, Corman, & Reichman, 2013; Mor et al., 2008). Items were scored dichotomously: 0 (never engaged in the specific risk behavior) or 1 (performed one or more episode of the specific risk behavior). Scores were calculated as the sum of the items of each submeasure: The higher the score, the less favorable the physical health risk behavior (on a scale from 0 to 8, α = .61), or the less favorable the sexual risk behavior (on a scale from 0 to 3, α = .67), as applicable.

Measures: Predictor Variables

Minority Stressors

Internalized homophobia was assessed using the Hebrew version of the LGB Self-Acceptance Questionnaire (Elizur & Mintzer, 2003). This comprised 10 questions concerning the respondents’ degree of acceptance of their sexual orientation, with responses on a five-point scale ranging from 1 (not at all) to 5 (very much). Elizur and Mintzer (2003) demonstrated convergent validity and very good internal consistency reliability (α = .81). The present study established a reliability of α = .82. Scores were calculated as the mean of the index items: The higher the score, the greater the internalized homophobia.

Level of outness (the degree to which the LGB individual has disclosed his or her sexual orientation to others) was assessed by seven items establishing the extent to which the subject had “come out” to close friends; family members; and friends and faculty or teachers at school, university, army, and work, respectively (adapted from D’Augelli et al., 2005), on a five-point scale ranging from 1 (no one) to 5 (all). D’Augelli et al. (2005) reported a reliability of α = .80 for the original scale. The present study established a reliability of α = .78. Scores were calculated as the mean of items: The higher the score, the greater the level of outness.

Distal and proximal antigay harassment were assessed by means of the adjusted Gay Harassment Scale (adapted from D’Augelli & Grossman, 2006). Participants were asked about the frequency of incidents of LGB-related verbal, physical, or sexual harassment; bullying; boycotts; or outings that they had witnessed or that happened to people they know (Distal Antigay Harassment Scale) over the previous year. Responses were rated on a four-point Likert-type scale, ranging from 1 (never) to 4 (three times or more). They were then presented with the same questions in relation to such incidents that they had experienced personally in the previous year (Proximal Antigay Harassment Scale). This study established a reliability of α = .71 for distal antigay harassment and α = .82 for proximal antigay harassment. Scores were calculated as the mean of the items constituting each scale: The higher the score, the more antigay harassment was experienced.

Coping Resources

Support by family and friends was rated on a 13-item scale (Abbey, Abramis, & Caplan, 1985) that measured perceived social support or social undermining from the individual’s close family and friends. Items were rated on a five-point Likert-type scale (ranging from 1 = not at all to 5 = a great deal). The Hebrew-language version of these scales was used in a study of LGB youths and young adults in Israel (Shilo & Savaya, 2011) that reported a good construct validity and a reliability of α = .87 and α = .84 for the support of family and support of friends, respectively. This study established a reliability of α = .92 for social support from friends and α = .88 for social support from family. Scores were calculated as the mean of the constituent items: The higher the score, the greater the support from each support provider. These items were completed by both LGB and heterosexual participants.

Connectedness to LGB community was assessed by means of an eight-item questionnaire (Shilo & Savaya, 2011), relating to three key social activities available to LGB youths and young adults in Israel: LGB social groups, Internet forums, and parties aimed at young LGBs. In each case, participants were asked to rate their social contact on a five-point scale, ranging from 1 (never) to 5 (usually). Shilo and Savaya (2011) reported a good construct validity and a reliability of α = .79. This study established a reliability of α = .76. Scores were calculated as the mean of the constituent items: The higher the score, the higher the individual’s connectedness to the LGB community.

Statistical Analysis

Data were analyzed and cleaned; no missing data were detected. Physical health outcomes were dichotomized on the basis of their median scores (range of physical risk behavior: 0 to 8, median = 2; range of sexual risk behavior: 0 to 3, median = 1); univariate analysis was carried out by chi-square test for categorical variables or an independent Student’s t test for continuous variables for normally distributed variables using SPSS for Windows software (version 18). Independent variables were included in a regression model (linear and logistic regression, for mental and physical health outcomes, respectively), which was carried out to examine predictors of mental and physical health outcomes, controlled for place of residency. Analyses were performed in two stages: First, we analyzed the entire sample for differences between LGB and heterosexual participants with regard to mental and physical health, and for determinants predicting poor mental health and health risk behaviors. In this stage, independent variables included demographics (age, gender, sexual orientation) and coping resources (support of family and friends). Second, we analyzed only the LGB participants for determinants predicting poor mental health and health risk behaviors. In this stage, independent variables included demographics (age, gender), minority stressors (internalized homophobia, level of outness, distal and proximal antigay harassment), and coping resources (support of family and friends, connectedness to LGB community) predicting mental health and health risk behavior outcomes. In addition, mental health variables were included as predictors for risk behavior outcomes. To control for Type I errors in regression models, where multiple independent variables were tested, we used Bonferroni correction (critical alpha = .05/7 = .007 for multivariate analysis of the entire sample; critical alpha = .05/11 = .004 for multivariate analysis of LGB participants’ sample).

Results

Physical and Mental Health among LGB Participants Compared with Heterosexuals

The comparison between the mental and physical health of LGB participants and their heterosexual counterparts is shown in Table 2. It shows that LGB participants exhibited greater mental distress (M = 67.33 versus 62.10, t(950) = 3.47, p < .001) and lower levels of well-being (M = 45.91 versus 50.15, t(950) = –4.63, p < .001), and engaged in more physical (M = 2.96 versus 2.62, t(950) = 2.84, p < .05) and sexual risk behaviors (M = 1.12 versus 1.39, t(950) = 3.53, p < .001) compared with heterosexuals. In addition, support of family (M = 3.21 versus 3.98, t(950) = –2.20, p = .01) and friends (M = 4.01 versus 4.23, t(950) = –2.67, p < .001) was significantly lower among LGB participants compared to heterosexuals.

Table 2:

Health Outcomes among Study Participants, by Sexual Orientation (N = 952)

 Heterosexuals (n = 267)
 
LGBs (n = 685)
 
 
Variable M SD M SD p 
Mental distress 62.10 20.38 67.33 21.87 .00 
Well-being 50.15 12.32 45.91 12.80 .00 
Physical risk behavior 2.62 2.33 3.18 2.96 .02 
Sexual risk behavior 1.39 0.85 1.65 1.12 .00 
Family support 3.98 0.75 3.21 0.81 .01 
Friends’ support 4.23 0.55 4.01 0.51 .00 
 Heterosexuals (n = 267)
 
LGBs (n = 685)
 
 
Variable M SD M SD p 
Mental distress 62.10 20.38 67.33 21.87 .00 
Well-being 50.15 12.32 45.91 12.80 .00 
Physical risk behavior 2.62 2.33 3.18 2.96 .02 
Sexual risk behavior 1.39 0.85 1.65 1.12 .00 
Family support 3.98 0.75 3.21 0.81 .01 
Friends’ support 4.23 0.55 4.01 0.51 .00 

Note: LGBs = lesbians, gays, and bisexuals.

Multivariate regression identifying variables predicting mental and physical health in the entire sample (see Table 3) demonstrated that being female (B = 6.47, p < .007) and LGB (B = 8.22, p < .007) was independently associated with high levels of mental distress. Being adolescent (B = 3.83, p < .007) and LGB (B = –5.5, p < .007) was also independently associated with lower levels of well-being. Higher levels of family and friends’ support were associated with lower levels of mental distress and higher levels of well-being. Being a young adult (B = –0.92, p < .007), LGB (B = 0.42, p < .007), with higher levels of friends’ support, and mentally distressed was independently associated with physical risk behavior; being a young adult (B = –1.32, p < .007), gay or bisexual male (gender: B = –0.81, LGB: B = –0.19, p < .007), with higher levels of friends’ support, higher levels of mental distress, and lower levels of well-being was independently associated with sexual risk behavior. LGB participants were 1.5 times more likely than heterosexuals to engage in physical risk behaviors and 1.2 times more likely to engage in sexual risk behaviors.

Table 3:

Multivariate Analysis of Determinants Influencing Mental and Physical Health Outcomes (N = 952)

 Mental Health
 
Physical Health
 
 Mental Distress
 
Well-being
 
Physical Risk Behavior
 
Sexual Risk Behavior
 
Variable B SE B B SE B B OR (95% CI) B OR (95% CI) 
Age ≤ 18 –2.42 1.45 3.83 0.86 0.92 0.40 (0.290.55) –1.32 0.27 (0.180.40) 
Female 6.47 1.41 –0.85 0.84 0.03 1.03 (0.76–1.40) 0.81 0.44 (0.310.63) 
LGB 8.22 1.57 5.5 0.93 0.42 1.47 (1.042.06) 0.19 1.21 (1.112.38) 
Family support 6.86 0.85 2.83 0.50 0.13 1.40 (0.94–1.43) 0.17 1.19 (0.96–1.48) 
Friends’ support 10.19 1.25 7.88 0.74 0.48 1.62 (1.222.17) 0.58 1.80 (1.292.51) 
Mental distress     0.03 1.03 (1.021.06) 0.02 1.02 (1.011.03) 
Well-being     –0.01 0.99 (0.98–1.01) 0.04 0.97 (0.950.98) 
R2 .186 .191 .141 .235 
 Mental Health
 
Physical Health
 
 Mental Distress
 
Well-being
 
Physical Risk Behavior
 
Sexual Risk Behavior
 
Variable B SE B B SE B B OR (95% CI) B OR (95% CI) 
Age ≤ 18 –2.42 1.45 3.83 0.86 0.92 0.40 (0.290.55) –1.32 0.27 (0.180.40) 
Female 6.47 1.41 –0.85 0.84 0.03 1.03 (0.76–1.40) 0.81 0.44 (0.310.63) 
LGB 8.22 1.57 5.5 0.93 0.42 1.47 (1.042.06) 0.19 1.21 (1.112.38) 
Family support 6.86 0.85 2.83 0.50 0.13 1.40 (0.94–1.43) 0.17 1.19 (0.96–1.48) 
Friends’ support 10.19 1.25 7.88 0.74 0.48 1.62 (1.222.17) 0.58 1.80 (1.292.51) 
Mental distress     0.03 1.03 (1.021.06) 0.02 1.02 (1.011.03) 
Well-being     –0.01 0.99 (0.98–1.01) 0.04 0.97 (0.950.98) 
R2 .186 .191 .141 .235 

Notes: Boldface indicates significant level after Bonferroni correction (p ≤ .007); CI = confidence interval; LGB = lesbian, gay, and bisexual.

Predictors of Mental and Physical Health among Young LGB Participants

Females and adolescents reported more mental distress compared with males and young adults (see Table 4). All minority stressors and coping resources were significantly correlated in the expected direction with mental distress and well-being—except for level of outness and distal antigay harassment, which were not significantly correlated with well-being.

Table 4:

Characteristics of Mental and Physical Health among Lesbian, Gay, and Bisexual Participants (n = 685)

 Mental Distress
 
Well-being
 
Physical Risk Behavior
 
Sexual Risk Behavior
 
Variable M (SD) p M (SD) p High Levelsan (%) Low Levelsbn (%) p High Levelsan (%) Low Levelsbn (%) p 
None (n = 685) 67.33 (21.87)  45.91 (12.80)  412 (60) 273 (40)  248 (36) 437 (64)  
Male 65.09 (21.78)  45.83 (12.91)  262 (60) 40 (176)  41 (178) 59 (260)  
Female 71.34 (21.49) .000d 46.05 (12.62) .828d 150 (61) 39 (97) .718c 27 (70) 73 (177) .000c 
Adolescent ( ≤ 18) 70.83 (22.56)  46.20 (12.63)  110 (47) 53 (125)  17 (40) 83 (192)  
Young adults (>18) 65.50 (21.29) .002d 45.75 (12.89) .667d 302 (67) 33 (144) .000c 46 (208) 54 (245) .000c 
 Pearson’s r  Pearson’s r        
Intern. homophobia .21 .000e –.18 .000e 1.56 (.51) 1.51 (.52) .156d 1.47 (.51) 1.56 (.54) .038d 
Level of outness –.10 .007e .06 .098e 3.49 (1.07) 2.96 (1.09) .000d 3.76 (.93) 3.01 (1.11) .000d 
Distal harassment .17 .000e –.05 .203e 11.12 (4.07) 9.81 (3.99) .000d 12.03 (4.24) 9.92 (3.85) .000d 
Proximal harassment .14 .000e –.09 .020e 7.99 (2.74) 6.97 (2.39) .000d 8.73 (2.54) 6.97 (2.49) .000d 
 –.33 .000e .24 .000e 3.89 (.82) 3.82 (.78) .259d 3.95 (.80) 3.81 (.81) .029d 
Family support –.31 .000e .36 .000e 4.26 (.52) 4.17 (.58) .039d 4.31 (.46) 4.18 (.59) .003d 
Friends’ support –.10 .009e .14 .000e 2.54 (.76) 2.58 (.87) .478d 2.58 (.78) 2.55 (.83) .539 
LGB connectedness – – –.69 .000e 69.89 (21.43) 63.51 (22.02) .000d 72.42 (19.82) 64.54 (22.49) .000d 
Mental distress –.69 .000e – – 44.80 (12.87) 47.64 (12.52) .004d 42.02 (11.33) 48.15 (13.10) .000d 
 Mental Distress
 
Well-being
 
Physical Risk Behavior
 
Sexual Risk Behavior
 
Variable M (SD) p M (SD) p High Levelsan (%) Low Levelsbn (%) p High Levelsan (%) Low Levelsbn (%) p 
None (n = 685) 67.33 (21.87)  45.91 (12.80)  412 (60) 273 (40)  248 (36) 437 (64)  
Male 65.09 (21.78)  45.83 (12.91)  262 (60) 40 (176)  41 (178) 59 (260)  
Female 71.34 (21.49) .000d 46.05 (12.62) .828d 150 (61) 39 (97) .718c 27 (70) 73 (177) .000c 
Adolescent ( ≤ 18) 70.83 (22.56)  46.20 (12.63)  110 (47) 53 (125)  17 (40) 83 (192)  
Young adults (>18) 65.50 (21.29) .002d 45.75 (12.89) .667d 302 (67) 33 (144) .000c 46 (208) 54 (245) .000c 
 Pearson’s r  Pearson’s r        
Intern. homophobia .21 .000e –.18 .000e 1.56 (.51) 1.51 (.52) .156d 1.47 (.51) 1.56 (.54) .038d 
Level of outness –.10 .007e .06 .098e 3.49 (1.07) 2.96 (1.09) .000d 3.76 (.93) 3.01 (1.11) .000d 
Distal harassment .17 .000e –.05 .203e 11.12 (4.07) 9.81 (3.99) .000d 12.03 (4.24) 9.92 (3.85) .000d 
Proximal harassment .14 .000e –.09 .020e 7.99 (2.74) 6.97 (2.39) .000d 8.73 (2.54) 6.97 (2.49) .000d 
 –.33 .000e .24 .000e 3.89 (.82) 3.82 (.78) .259d 3.95 (.80) 3.81 (.81) .029d 
Family support –.31 .000e .36 .000e 4.26 (.52) 4.17 (.58) .039d 4.31 (.46) 4.18 (.59) .003d 
Friends’ support –.10 .009e .14 .000e 2.54 (.76) 2.58 (.87) .478d 2.58 (.78) 2.55 (.83) .539 
LGB connectedness – – –.69 .000e 69.89 (21.43) 63.51 (22.02) .000d 72.42 (19.82) 64.54 (22.49) .000d 
Mental distress –.69 .000e – – 44.80 (12.87) 47.64 (12.52) .004d 42.02 (11.33) 48.15 (13.10) .000d 

Note: Intern. = Internalized; LGB = lesbian, gay, and bisexual.

aAbove or equal to the median score.

bBelow the median score.

cχ2, % (N).

dStudent’s t test, M (SD).

ePearson’s r.

With regard to physical health of all 685 LGB participants, 412 (60 percent) engaged in high levels of physical risk behaviors during the previous six months. Typically, these were young adults (67 percent) with higher levels of outness, who had experienced greater distal and proximal antigay harassment, greater support from friends, and high levels of mental distress and low levels of well-being. With regard to sexual risk behaviors, 248 respondents (35.6 percent) were engaged in high levels of sexual risk behaviors during the previous six months. These were commonly young adult (46 percent) males (41 percent), with higher levels of internalized homophobia and higher levels of outness, who had experienced greater distal and proximal antigay harassment, greater support from family and friends, and exhibited lower levels of well-being and higher levels of mental distress.

Multivariate regression analyses identifying variables predicting higher levels of physical and mental health found that being a female young adult had a significant association with mental distress (B = 6.18, p < .004). In addition, almost all minority stressors (high levels of internalized homophobia, distal and proximal antigay harassment) and coping resources (lower levels of support from friends and family) were significantly associated with mental distress (see Table 5). Low levels of internalized homophobia, high levels of outness, and low levels of coping resources (support from friends and family) were found to predict lower levels of well-being among young LGB participants. From all predictors of mental health outcomes, friends’ support had the strongest effect on both mental distress (B = –8.98, p < .001) and well-being (B = 7.62, p < .001).

Table 5:

Multivariate Analysis of Variables Predicting Mental and Physical Health Outcomes among LGB Participants (n = 685)

 Mental Health
 
Physical Health
 
 Mental Distress
 
Well-being
 
Physical Risk Behavior
 
Sexual Risk Behavior
 
Variable B SE B SE B OR (95% CI) B OR (95% CI) 
Age ≤ 18 0.9 1.82 2.09 1.09 –0.60 0.85 (0.36–0.86) –1.07 0.34 (0.21–0.57) 
Female 6.18 1.67 –0.46 –0.27 0.77 (0.52–1.12) –0.77 0.47 (0.30–0.72) 
Internalized homophobia 8.04 1.64 -3.69 0.99 0.07 1.07 (0.73–1.55) –0.05 0.95 (0.62–1.47) 
Level of outness –2.08 0.89 1.5 0.54 0.3 1.35 (1.10–1.65) 0.34 1.47 (1.17–1.84) 
Distal harassment 0.65 0.24 –0.07 0.14 0.03 1.03 (0.97–1.08) 0.07 1.08 (1.01–1.14) 
Proximal harassment 0.48 1.05 –0.16 0.22 0.11 1.12 (1.03–1.22) 0.19 1.21 (1.09–1.32) 
Family support –6.06 1.02 2.58 0.61 0.15 1.16 (0.92–1.47) 0.13 1.14 (0.87–1.49) 
Friends’ support –8.98 1.52 7.62 0.91 0.21 1.23 (0.86–1.77) 0.68 1.97 (1.23–3.07) 
LGB connectedness –2.23 1.36 0.6 –0.20 0.80 (0.64–1.01) 0.30 1.36 (1.03–1.42) 
Mental distress     0.49 1.64 (1.01–1.73) 0.01 1.01 (0.99–1.02) 
Well-being     –0.01 0.99 (0.98–1.02) –0.04 0.96 (0.94–0.98) 
R2 .233 .193 .163 .365 
 Mental Health
 
Physical Health
 
 Mental Distress
 
Well-being
 
Physical Risk Behavior
 
Sexual Risk Behavior
 
Variable B SE B SE B OR (95% CI) B OR (95% CI) 
Age ≤ 18 0.9 1.82 2.09 1.09 –0.60 0.85 (0.36–0.86) –1.07 0.34 (0.21–0.57) 
Female 6.18 1.67 –0.46 –0.27 0.77 (0.52–1.12) –0.77 0.47 (0.30–0.72) 
Internalized homophobia 8.04 1.64 -3.69 0.99 0.07 1.07 (0.73–1.55) –0.05 0.95 (0.62–1.47) 
Level of outness –2.08 0.89 1.5 0.54 0.3 1.35 (1.10–1.65) 0.34 1.47 (1.17–1.84) 
Distal harassment 0.65 0.24 –0.07 0.14 0.03 1.03 (0.97–1.08) 0.07 1.08 (1.01–1.14) 
Proximal harassment 0.48 1.05 –0.16 0.22 0.11 1.12 (1.03–1.22) 0.19 1.21 (1.09–1.32) 
Family support –6.06 1.02 2.58 0.61 0.15 1.16 (0.92–1.47) 0.13 1.14 (0.87–1.49) 
Friends’ support –8.98 1.52 7.62 0.91 0.21 1.23 (0.86–1.77) 0.68 1.97 (1.23–3.07) 
LGB connectedness –2.23 1.36 0.6 –0.20 0.80 (0.64–1.01) 0.30 1.36 (1.03–1.42) 
Mental distress     0.49 1.64 (1.01–1.73) 0.01 1.01 (0.99–1.02) 
Well-being     –0.01 0.99 (0.98–1.02) –0.04 0.96 (0.94–0.98) 
R2 .233 .193 .163 .365 

Notes: Boldface indicates significant level after Bonferroni correction (p ≤ .004); LGB = lesbian, gay, and bisexual.

Young adults and participants with high levels of outness, as well as those exposed to antigay harassment, were more likely to engage in physical risk behaviors. As for sexual risk behavior, being a male young adult, feeling comfortable in sharing one’s sexual orientation, experiencing more proximal antigay harassment, receiving greater support from friends, and tighter connectedness to the LGB community were all found to predict sexual health risk behavior. With regard to the association between mental health and physical risk behavior, higher levels of mental distress were found to predict physical risk behaviors (B = 0.49, p < .004), while low levels of well-being determinants were found to predict sexual risk behaviors (B = –0.04, p < .004).

Discussion

This study’s findings, that LGB youths and young adults experience poorer mental and physical health than heterosexuals and are more frequently involved in physical and sexual risk behaviors, are consistent with findings of previous studies conducted in Western societies, and with minority stress theory (Kertzner et al., 2009; Russell & Joyner, 2001). Although our LGB sample comparison to heterosexuals was limited, we found that LGB participants had significantly lower levels of family and friends’ support. Other stressors that may indeed affect mental and physical health (for example, stressful life events) were not tested in the current study. However, these general stressors were not found in previous research to exemplify LGB people when compared with heterosexuals (see, for example, Meyer, Schwartz, & Frost, 2008).

We found that minority stressors had a negative association and coping resources had a positive association on both mental and physical health of LGB youths and young adults. Experiences of antigay harassment, which in the past have been found to affect mental health in LGBs (Kertzner et al., 2009), were found in this study to be associated with physical health, as well: Firsthand (proximal) experiences of antigay harassment were associated positively with mental distress, and with increased physical and sexual risk behaviors, whereas personal acquaintance with LGB individuals who have been harassed because of their sexual orientation (distal encounters) was associated with mental distress. High levels of internalized homophobia and low levels of support from friends and family were associated with high levels of mental distress and low levels of well-being. In addition, high levels of connectedness to the LGB community were associated with high levels of well-being. These findings provide further evidence of the key role played in the mental health of sexual minorities by support providers in the lives of young LGBs, of the importance of accepting one’s sexual orientation, and of the value of social connections to the LGB community (Shilo & Savaya, 2011).

Our findings demonstrate that minority stressors and coping resources were associated with physical health in a different manner than that of mental health. Contrary to our initial hypotheses, high levels of outness were associated with both physical and sexual risk behaviors, and connectedness to the LGB community and friends’ support were associated with increased incidence of sexual risk behavior. These findings suggest that coping resources are at once both protective and risk factors for young LGBs. It is possible that the social sense of security due to belonging to a wider LGB community once young LGBs “come out,” coupled with the support of family and friends, has a positive effect on their mental health, while simultaneously exposing them to greater risk of sexual behaviors. The greater sexual risk is probably a result of the increased access of young LGBs to gay- or lesbian-oriented venues, which provide the opportunity to meet more sexual partners, and of adopting similarly promiscuous patterns of behavior. These findings expand our understanding of the complex effects of minority stressors on sexual minorities’ health, whereas previous research has focused almost solely on mental health outcomes (Kertzner et al., 2009; Meyer et al., 2008).

The riskier physical and sexual behaviors among LGB youths and young adults, as compared with heterosexuals, may be partially explained by their state of mental health: In this study, physical risk behavior was associated with high levels of mental distress, and sexual risk behavior was associated with lower levels of well-being. Minority stress perspective views sexual minority–related stressors as affecting both mental and physical health (Meyer, 2007). Our findings suggest that these stressors represent vulnerability for the subsequent development of mental distress. Studies of co-occurring (or “syndemic”) psychological problems show that there is an additional risk with psychological distress with respect to sexual risk behavior in gay men (Mustanski, Garofalo, Herrick, & Donenberg, 2007; Safran, Blashill, & O’Cleirigh, 2011). Depressive symptoms may adversely affect self-esteem, self-assertion, and self-protective behaviors, and may contribute to higher rates of substance use and other risk-taking behaviors among young men who have sex with men (Salomon, Mimiaga, & Husnik, 2009).

Among all participants, women exhibited higher levels of mental distress whereas men were more often engaged in physical and sexual risk behaviors, in line with other studies (Courtenay, 2000; King et al., 2008). These findings may be explained by general gender differences in risk perception: Men tend to take greater risks than women (Mahalik, Burns, & Syzdek, 2007). Young adulthood was found to be a risk factor for physical and sexual risk behaviors compared with adolescence—both in the entire study sample and among LGB participants. One possible explanation for those findings is that accessibility to physical risk behaviors (such as drugs and alcohol) as well as opportunities for sexual risk behaviors (sexual intercourse and sexual lifestyle) increase with age. LGB young adults were more likely than adolescents to engage in physical risk behaviors, suggesting that physical health risk behaviors increase with age at a higher rate in the LGB community. In addition, being adolescent (under age 18) was not found to predict lower levels of mental health among young LGBs. These findings support a recently contested standpoint for the deficit model of LGB youth development (Savin-Williams, 2005) and provide further evidence to findings that LGB adolescents are less vulnerable to mental health issues than minority stress and developmental theories of LGBs suggest.

Interpretation of these findings is limited by various factors. First, the study was based on a convenience sample, which may not be representative of all LGB youths and young adults in Israel. Second, the cross-sectional design limits the conclusions that may be drawn regarding causality. To allow temporality, the time frame of physical and sexual behaviors and of mental health feelings was given prominence in the questionnaire. Third, all the findings are based on self-reporting, which is susceptible to reporting bias, especially with regard to sensitive and intimate aspects such as sexual practices and substance use. To minimize this bias, participants were asked to complete the data anonymously and only for the previous six months. Finally, targeting youths and young adults allowed us to assess only physical risk behaviors, rather than confirmed medical health outcomes. This young population is normally healthy, which limits our ability to detect actual medical deterioration in physical health. Our findings suggest that the effect of minority stressors on physical health increases with age, so additional longitudinal studies are needed to assess the association of such changes with age. Despite these limitations, this study does provide a snapshot of mental and physical health in a non-Christian Western society, allowing us to probe the effects of social stressors on mental and physical health and to link the findings with gender, age, and sexual orientation.

The findings of this study have several implications for practice and research. Social workers and other professionals working with LGB youths and young adults can use the minority stress model as an assessment tool, underscoring antigay harassment experiences, internalized homophobia, and the lack of social support as mental health risk factors that have a negative impact on physical and sexual risk behaviors. Most of the literature on social work practice with LGB youths and young adults adopts coming out models that emphasize the importance of a healthy coming out process and its positive effects on mental health (Mallon, 2008; Morrow & Messinger, 2006). This study suggests that while coming out as a sexual minority may indeed serve as a protective factor for mental health, it also exposes adolescents and young adults to physical and sexual risk behaviors. Social workers and health professionals working with LGB youths and young adults should help young LGB individuals navigate their social lives and behaviors during and after coming out to friends and family, and to be aware of the physical and sexual risks related to the unique subcultural climate of the LGB community. In addition, findings of this research emphasize the importance of assessing both physical and mental health components when working with LGB youths and young adults. Social workers and health educators should exploit the fact that young LGBs who have “come out” are more likely to frequent LGB-related social venues and to use these venues to advertise health announcements and safe-sex messages tailored to sexual minorities of young ages. As LGB people in developed countries share a similar subculture, the results of this study may be used by any health professionals who wish to perform risk reduction interventions in the LGB community.

References

Abbey
A.
Abramis
D. J.
Caplan
R. D.
Effects of different sources of social support and social conflict on emotional well-being
Basic and Applied Social Psychology
 , 
1985
, vol. 
6
 
2
(pg. 
111
-
129
)
Central Bureau of Statistics (Israel)
Usage of technology among Israeli population over 20 years old [in Hebrew]
 , 
2009
 
Cochran
S. D.
Sullivan
J. G.
Mays
V. M.
Prevalence of mental disorders, psychological distress, and mental health services use among lesbian, gay, and bisexual adults in the United States
Journal of Consulting and Clinical Psychology
 , 
2003
, vol. 
71
 
1
(pg. 
53
-
61
)
Courtenay
W. H.
Behavioral factors associated with disease, injury, and death among men: Evidence and implications for prevention
Journal of Men’s Studies
 , 
2000
, vol. 
9
 
1
(pg. 
81
-
142
)
Dar
Y.
Kimhi
S.
Military service and self-perceived maturation among Israeli youth
Journal of Youth and Adolescence
 , 
2001
, vol. 
30
 (pg. 
427
-
448
)
D’Augelli
A. R.
Omoto
A. M.
Kurtzman
H. S.
Developmental and contextual factors and mental health among lesbian, gay, and bisexual youths
Sexual orientation and mental health
 , 
2006
Washington, DC
American Psychological Association
(pg. 
37
-
53
)
D’Augelli
A. R.
Grossman
A. H.
Researching lesbian, gay, and bisexual youth: Conceptual, practical, and ethical issues
Journal of Gay & Lesbian Issues in Education
 , 
2006
, vol. 
3
 
2–3
(pg. 
35
-
56
)
D’Augelli
A. R.
Grossman
A. H.
Salter
N. P.
Starks
M. T.
Vasey
J. J.
Sinclair
K. O.
Predicting the suicide attempts of lesbian, gay, and bisexual youth
Suicide and Life-Threatening Behavior
 , 
2005
, vol. 
35
 (pg. 
646
-
660
)
Eaton
D. K.
Kann
L.
Kinchen
S.
Shanklin
S.
Flint
K. H.
Youth risk behavior surveillance—United States, 2011
Surveillance Summaries
 , 
2012
, vol. 
61
 
4
(pg. 
1
-
162
)
Elizur
Y.
Mintzer
A.
Gay males’ relationship quality: The roles of attachment security, gay identity, social support, and income
Personal Relationships
 , 
2003
, vol. 
10
 (pg. 
411
-
435
)
Everett
S.
Corman
H.
Reichman
N. E.
Effects of overweight on risky sexual behavior of adolescent girls
Economic Inquiry
 , 
2013
, vol. 
51
 (pg. 
605
-
619
)
Florian
V.
Drori
J.
The mental health index: Psychometric characteristics and normative data in Israeli population
Psychologia
 , 
1990
(pg. 
26
-
35
(Hebrew)
Harte
C. B.
Meston
C. M.
Association between smoking cessation and sexual health in men
BJU International
 , 
2012
, vol. 
109
 (pg. 
888
-
896
)
Hass
S. A.
Schefer
D. R.
Kornienko
O.
Health and the structure of adolescent social networks
Journal of Health and Social Behavior
 , 
2010
, vol. 
51
 (pg. 
424
-
439
)
Horvath
K. J.
Remafedi
G.
Rosser
B.R.S.
Sexual risk taking among young Internet-using men who have sex with men
American Journal of Public Health
 , 
2008
, vol. 
98
 (pg. 
1059
-
1067
)
Ingram
R. E.
Price
J. M.
Vulnerability to psychopathology: Risk across the lifespan
 , 
2001
New York
Guilford Press
Kertzner
R. M.
Meyer
I. H.
Frost
D. M.
Stirratt
M. J.
Social and psychological well-being in lesbians, gay men, and bisexuals: The effects of race, gender, age, and sexual identity
American Journal of Orthopsychiatry
 , 
2009
, vol. 
79
 (pg. 
500
-
510
)
King
M.
Semlyen
J.
Tai
S. S.
Killaspy
H.
Osborn
D.
Popelyuk
D.
Nazareth
I.
A systematic review of mental disorder, suicide, and deliberate self-harm in lesbian, gay and bisexual people
BMC Psychiatry
 , 
2008
, vol. 
8
 pg. 
70
 
Kolbe
L.
Kann
L.
Patterson
B.
Wechsler
H.
Osorio
J.
Collins
J.
Enabling the nation’s schools to help prevent heart disease, stroke, cancer, COPD, diabetes, and other serious health problems
Public Health Reports
 , 
2004
, vol. 
119
 (pg. 
286
-
302
)
Mahalik
J. R.
Burns
S. M.
Syzdek
M.
Masculinity and perceived normative health behaviors as predictors of men’s health behaviors
Social Science & Medicine
 , 
2007
, vol. 
64
 (pg. 
2201
-
2209
)
Mallon
G. P.
Social work practice with LGBT people
 , 
2008
New York
Routledge
Meyer
I. H.
Prejudice, social stress, and mental health in LGB population: Conceptual issues and research evidence
Psychological Bulletin
 , 
2003
, vol. 
129
 (pg. 
674
-
697
)
Meyer
I. H.
Meyer
I. H.
Northridge
M. E.
Prejudice and discrimination as social stressors
The health of sexual minorities
 , 
2007
New York
Springer
(pg. 
242
-
267
)
Meyer
I. H.
Schwartz
S.
Frost
D. M.
Social patterning of stress and coping: Does disadvantaged social statuses confer excess exposure and fewer coping resources?
Social Science & Medicine
 , 
2008
, vol. 
67
 (pg. 
368
-
379
)
Meyer
I. H.
Wilson
P. A.
Sampling lesbian, gay, and bisexual populations
Journal of Counseling Psychology
 , 
2009
, vol. 
56
 (pg. 
23
-
31
)
Mor
Z.
Dan
M.
Knowledge, attitudes, sexual practice and HIV prevalence in men who have sex in Tel Aviv, Israel
Sexually Transmitted Infection
 , 
2012
, vol. 
88
 (pg. 
574
-
580
)
Mor
Z.
Davidovich
U.
Bessudu-Manor
N.
McFarlane
M.
High-risk behavior in steady and in casual relationships among MSM in Israel
Sexually Transmitted Infection
 , 
2012
, vol. 
87
 (pg. 
532
-
537
)
Mor
Z.
Davidovich
U.
McFarlane
M.
Feldstein
G.
Gay men who engage in substance use and sexual risk behaviour: A dual-risk group with unique characteristics
International Journal of STD & AIDS
 , 
2008
, vol. 
19
 (pg. 
698
-
703
)
Morrow
D. F.
Messinger
L.
Sexual orientation and gender expression in social work practice
 , 
2006
New York
Columbia University Press
Mustanski
B.
Garofalo
R.
Herrick
A.
Donenberg
G.
Psychosocial health problems increase risk for HIV urban young men who have sex with men: Preliminary evidence of a syndemic in need of attention
Annals of Behavioral Medicine
 , 
2007
, vol. 
34
 (pg. 
37
-
45
)
Rivers
I.
D’Augelli
A. R.
D’Augelli
A. R.
Patterson
C. J.
The victimization of lesbian, gay, and bisexual youths
Lesbian, gay, and bisexual identities and youth: Psychological perspectives
 , 
2001
New York
Oxford University Press
(pg. 
199
-
223
)
Russell
S. T.
Joyner
K.
Adolescent sexual orientation and suicide risk: Evidence from a national study
American Journal of Public Health
 , 
2001
, vol. 
91
 (pg. 
1276
-
1281
)
Safran
S. A.
Blashill
A. J.
O’Cleirigh
C. M.
Promoting the sexual health of MSM in the context of comorbid mental health problems
AIDS and Behavior
 , 
2011
, vol. 
15
 (pg. 
S30
-
S40
)
Salomon
E. A.
Mimiaga
M. J.
Husnik
M. J.
Depressive symptoms, utilization of mental health care, substance use and sexual risk among young men who have sex with men in EPOLRE: Implications for the age-specific interventions
AIDS and Behavior
 , 
2009
, vol. 
13
 (pg. 
811
-
821
)
Savin-Williams
R. C.
The new gay teenager
 , 
2005
Cambridge, MA
Harvard University Press
Shilo
G.
Savaya
R.
Effects of social support, undermining and acceptance on mental health and sexual orientation milestones of LGB youth
Family Relations
 , 
2011
, vol. 
60
 (pg. 
318
-
330
)
Veit
C. T.
Ware
J. E.
The structure of psychological distress and well-being in general populations
Journal of Consultation Clinical Psychology
 , 
1983
, vol. 
51
 (pg. 
730
-
742
)
Wong
C. F.
Kipke
M. D.
Weiss
G.
McDavitt
B.
The impact of recent stressful experiences on HIV-risk related behaviors
Journal of Adolescence
 , 
2010
, vol. 
33
 (pg. 
463
-
475
)