Butch/Femme Differences in Substance Use and Abuse Among Young Lesbian and Bisexual Women: Examination and Potential Explanations

Abstract

The current study examined the role of gender atypical self-presentation on the alcohol, tobacco, and marijuana use, as well as symptoms of substance abuse, of an ethnically diverse sample of 76 young (ages 14 – 21 years) lesbian and bisexual women who were interviewed between 1993-95 in New York City. Even after controlling for age, sexual identity, and social desirability, young butch women reported drinking alcohol more frequently and in greater quantity, smoking more cigarettes, and using marijuana more frequently than young femme women. Experiences of gay-related stressful events, internalized homophobia, and emotional distress were found to largely account for the butch/femme differences in tobacco and marijuana use, but not in higher levels of alcohol use. Despite the small convenience sample, these findings suggest that intervention efforts to address the higher levels of substance use among young lesbian and bisexual women may increase effectiveness by also addressing experiences of gay-related stress and emotional distress of young butch women.

Introduction

Over the past decade, an extensive body of research has examined potential disparities in substance use and abuse between lesbian, gay, and bisexual individuals and their heterosexual peers. The results of these large population-based samples have consistently documented that lesbian and bisexually identified women (as well as women with same-sex attractions and same-sex sexual experience) report higher levels of alcohol, marijuana, and tobacco use and greater symptoms of substance abuse than heterosexual women (Bontempo & D’Augelli, 2002; Burgard, Cochran, & Mays, 2005; Drabble, Midanik, & Trocki, 2005; McCabe, Boyd, Hughes, & D’Arcy, 2003; Ridner, Frost, & LaJoie, 2006; Ziyadeh et al., 2007). This greater risk for substance use and abuse has been documented in adult women (Burgard et al., 2005; Drabble et al., 2005), college students (Eisenberg & Wechsler, 2003; McCabe et al., 2003; Ridner et al., 2006), and other adolescents (Russell, Driscoll, & Truong, 2002; Ziyadeh et al., 2007). However, despite their well-documented risk for substance use and abuse, not all lesbian and bisexual women report high levels of recent heavy alcohol use (15% - 52%), tobacco use (26% - 57%), and marijuana use (16% - 33%; Austin et al., 2004; Bontempo & D’Augelli, 2002; Burgard et al., 2005; Drabble et al., 2005; Eisenberg & Wechsler, 2003; McCabe et al., 2003; Ridner et al., 2006; Ziyadeh et al., 2007). The reasons why and the conditions under which some lesbian and bisexual women may or may not report high levels of substance use and abuse remain far less examined.

One factor that may explain which lesbian/bisexual women are at greater risk for substance use and abuse is gender atypicality. Lesbian and bisexual women are significantly more likely than heterosexual women to report that they were “tomboys” as children (Phillips & Over, 1995; Singh, Vidaurri, Zambarano, & Dabbs, 1999; Whitam & Mathy, 1991), and subsequently report more “butch” or androgynous personality and behavior as adults (Carlson & Baxter, 1984; Lippa, 2005; Singh et al. 1999). However, such masculine self-presentation is by no means universal, with many lesbian and bisexual women describing more traditionally feminine childhoods and more “femme” self-presentation as adults. Such butch/femme self-presentation has a long history in the lesbian and bisexual female community that continues today both in the U.S. and internationally (Whitam, Daskalos, Sobolewski, & Padilla, 1998). Although there are no definitive estimates of the prevalence of butch/femme self-presentation, given the widespread use of convenience samples, studies have found that between 33% and 85% of lesbian women self-identify as butch or femme (Brown, Finn, Cooke, & Breedlove, 2002; Levitt & Horne, 2002; Loulan, 1990; Weber, 1996). Further, a growing literature has begun to examine potential butch/femme differences in personality (Loulan, 1990; Singh et al., 1999), mate selection (Bailey, Kim, Hills, & Linsenmeier, 1997; Levitt & Hiestand, 2005; Smith & Stillman, 2002), biological differences (Brown et al., 2002; Pearcey et al., 1996; Singh et al., 1999), and the coming-out process (Levitt & Horne, 2002; Rosario, Schrimshaw, Hunter, & Levy-Warren, In press).

There are a number of potential reasons to hypothesize that women who are more butch in their self-presentation would be more likely to use alcohol, tobacco, and marijuana with greater frequency and in greater quantity than women who are more femme in their self-presentation. First, heavy drinking, cigarette smoking, marijuana use, and substance abuse are more frequent among heterosexual males than heterosexual females (Grunbaum et al., 2002; Office of Applied Statistics, 2004; Wallace et al., 2003) and, as such, are typically considered more masculine behaviors. Therefore, it might be expected that butch lesbians would engage in higher levels of substance use because it would be consistent with their more masculine self-presentation in other domains (e.g., men’s clothing, short hair). Second, although gender atypical self-presentation has not always been associated with greater emotional distress (Skidmore, Linsenmeier, & Bailey, 2006), it has been suggested that lesbians with a more butch self-presentation may experience higher levels of gay-related stressors (e.g., victimization, ridicule) due to their greater visibility as lesbian (Levitt & Horne, 2002; Levitt & Hiestand, 2004; Wyss, 2004). Representative studies of high school students have found that victimization is much more common among LGB than heterosexual youths (2.9 times per year among lesbian and bisexual female youths as compared with 0.7 times for heterosexual girls, and 5.5 times per year among gay and bisexual male youths v. 1.1 times for heterosexual boys: Bontempo & D’Augelli, 2002). These high levels of gay-related stressful events may, in turn, lead young butch women to use alcohol, tobacco, and marijuana in order to cope. Indeed, several studies (including those based on representative samples) have found that gay-related stressful events are associated with high levels of substance use and abuse among lesbian and bisexual women (Bontempo & D’Augelli, 2002; Nawyn, Richman, Rospenda, Hughes, 2000; see Hughes & Eliason, 2002 for review), although not in all studies (Heffernan, 1998). In addition, gay-related stressors may place young butch lesbians at greater risk for emotional distress (e.g., Bontempo & D’Augelli, 2002). Symptoms of depression, anxiety, and conduct problems have been found to be associated with higher levels of substance use and abuse among large representative samples of lesbian and bisexual women (Corliss, Grella, Mays, & Cochran, 2006; Ziyadeh et al., 2007; see Hughes & Eliason, 2002 for review).

Alternatively, there is reason to suggest that women with a more femme self-presentation may report higher levels of alcohol, tobacco and marijuana use than more butch women. Although butch women may experience more gay-related stressful events, femme women may have more difficulty resolving internalized homophobia (Levitt, Gerrish, & Hiestand, 2003). Specifically, femmes are more likely to “pass” or be perceived as heterosexual (Loulan, 1990), and therefore some femme women report that they had a difficult time acknowledging to themselves and others that they were lesbian (Levitt et al., 2003). Indeed, in an earlier report on this sample, femmes were less comfortable than butches with disclosing their sexuality to others (Rosario et al., in press). Furthermore, some research has suggested that femme women become aware of their sexual orientation at a later age than butch women (Levitt & Horne, 2002), implying that they have had less time to resolve their own internalized homophobia than butch women. Such unresolved negative attitudes toward their sexuality and discomfort with others knowing have been associated with both greater substance use and abuse (Amadio, 2006; DiPlacido, 1998) and more emotional distress (Lewis, Derlega, Clarke, & Kuang, 2006; Szymanski, Chung & Balsam, 2001), which, in turn, may lead to greater substance use (Corliss et al., 2006; Ziyadeh et al., 2007; see Hughes & Eliason, 2002 for review).

To date only a single study has examined potential butch/femme differences in substance use. The study failed to identify any significant butch/femme differences in either cigarette smoking or alcohol consumption (Singh et al., 1999). However, given the lack of research into this question and the numerous theoretical reasons to hypothesize differences (albeit, conflicting suggestions about the direction of these differences), further research is needed. Such research would provide important information about which young lesbian and bisexual women may be at risk for substance use and thereby suggest which women may be in most need of substance use prevention and treatment efforts. The current report builds on past research to examine potential butch/femme differences in substance use and symptoms of substance abuse and to investigate whether gay-related stressful events, internalized homophobia, or emotional distress may account for any observed butch/femme differences.

Method

Participants

Female and male youths, ages 14 to 21 years, were recruited from three gay-focused community-based organizations (85%) and two college student organizations (15%) in New York City. Meetings were held with the study investigators to introduce the study and invite the adolescents to participate in a study of lesbian, gay, and bisexual youths. In addition, youths were approached individually and invited to participate. These recruitment efforts were conducted by a gender and ethnically diverse study staff who were college educated and trained to follow a common recruitment procedure. Attempts were made to recruit every youth attending the recruitment site. We have estimated that approximately 80% of youths who were invited or attended a meeting participated in the study.

Of the 164 youths interviewed at Time 1, 49% were female. The current report focuses on these lesbian and bisexual female youths. Of the 81 young women interviewed, 1 was excluded because she was older than 21 years and 4 were excluded because they were heterosexual in both identity and lifetime sexual behavior. The resulting sample of 76 female youths reported a mean age of 18.4 years (SD = 1.6). Female youths were of Latino (38%), Black (36%), White (20%), and other (7%) ethnic backgrounds. Although 16% were recruited from college organizations, 32% were in college. Thirty-eight percent reported that they had a parent who received welfare, food stamps, or Medicaid. Youths self-identified as lesbian (67%), bisexual (32%), or other (1%).

Procedure

Voluntary and signed informed consent was provided by all youths. In addition, for youths under age 18 years, parental consent was waived by the Commissioner of Mental Health for New York State. The university and recruitment sites approved the study.

Youths participated in a 2- to 3-hour interviewer-administered questionnaire battery at Time 1, with follow-up assessments 6 and 12 months later. Interviews were conducted between fall 1993 and spring 1994, with follow-up through summer 1995. Interviewers were matched to the youths on gender. Interviews were conducted in a private room at each recruitment site at Time 1, with follow-up interviews conducted either at the recruitment site or another private location convenient for the youths. Youths received $30 at each interview. Only 4 (5%) female youths were lost to both follow-up assessments. Retention rates for the female youths were 87% at Time 2 and 88% at Time 3.

Measures of Substance Use and Abuse

The quantity of cigarette, alcohol, and marijuana use, as well as the frequency of alcohol and marijuana use, were assessed at all three assessments using the Alcohol and Drugs Schedule (ADS) which has demonstrated good test-retest reliability (Rosario, Hunter, & Gwadz, 1997). The ADS was adapted from several national survey measures of substance use among youths and it was tailored to include the street terms commonly used for these substances in New York City (see Rosario et al., 1997 for details). All questions assessed substance use in the past 3 months at Time 1, and within the past 6 months (since the last interview) at each subsequent assessment. The average quantity of cigarette use per day was assessed on a 7-point scale ranging from “Did not smoke cigarettes in the past [three/six] months” (0) to “About two packs or more per day (over 35 cigarettes)” (6). The quantity of alcohol and marijuana use was assessed as how many drinks youths typically have when they drink or how many joints they usually have when they use marijuana. Because data for quantity of alcohol and marijuana were positively skewed, the quantity of alcohol was computed as 0, 1, 2, or 3 or more drinks, and the quantity of marijuana was computed as 0, 1, or 2 or more joints. The frequency of alcohol and marijuana use was assessed as the number of times youths had any alcoholic beverage in the past three/six months and as the number of times they used marijuana in the past three/six months. Given the positively skewed nature of these data, the frequency of both alcohol and marijuana use was recoded such that 0 = 0 times, 1 = 1 – 6 times, and 2 = 7 or more times.

The ADS also included symptoms associated with abuse of alcohol or illicit drug abuse. These symptoms were assessed at all three assessments using an 11-item measure (e.g., Felt you needed or were dependent on alcohol and/or drugs) derived from the Diagnostic Interview Schedule for Children (National Institute of Mental Health, 1992). Items use a 5-point response scale from “not at all” (1) to “very often” (5). The mean of the items was computed as an index of substance abuse (α = .78 - .92 across all 3 assessment times for the female youths).

Measures of Butch and Femme

The classification of the young women as butch and femme was based on the composite of three types of items: 1) the youths’ self-identification as butch/femme, 2) the youth’s report of how other lesbians would identify her, and 3) the interviewer’s rating of the youth. With respect to self-identification, youths were asked at each assessment period three separate items regarding whether or not “you think of yourself or see yourself” as butch, as femme, or as androgynous. Youths responded yes or no to each item. These items have demonstrated strong test-retest reliability (Rosario et al., in press). Responses to the three items were combined such that young women who identified only as butch, or as butch and androgynous were classified as butch, whereas youths who identified as femme or as femme and androgynous were classified as femme. Youths who identified as butch and femme, as butch, femme, and androgynous, or only as androgynous were classified as androgynous. Finally, young women who did not identify as butch, femme, or androgynous were defined as “none of the above.” Therefore, a single variable was computed for each assessment period that classified young women as butch, femme, androgynous, or none of the above. The rationale for this classification system is available elsewhere (Rosario et al., in press).

Because we anticipated that some young women might be reluctant to self-identify as butch or femme, we also asked the youths to indicate how “on average other lesbians see you or react to you.” Three items assessed whether other lesbians would perceive her as butch, as femme, or as androgynous. These items have demonstrated strong test-retest reliability (Rosario et al., in press). Following the same classification system outlined above, we computed a single variable for each assessment period that classified the young women as butch, femme, androgynous, or none of the above. In addition, to further avoid reluctance to self-identify as butch/femme, we obtained interviewer ratings at Time 3 only. Interviewers rated each female youth on whether she would “be perceived by the average casual observer, such as people on the street” as butch, femme, or androgynous. Because the interviewer ratings were obtained at Time 3 only, no test-retest data were available.

These three measures of butch/femme were correlated (Contingency Coefficients, CC, = .56 - .67; Rosario et al., in press). Therefore, the three measures of butch/femme were combined into a single composite measure of butch/femme identity for each time period. Specifically, young women were classified as the maximum value reported across the measures (butch = 3, femme = 2, androgynous = 1, and none of the above = 0), resulting in a single composite measure of butch/femme for each assessment period. These three composite variables were also correlated over time (CC = .61 - .72). Therefore, we combined the three composite butch/femme variables at Time 1, 2, and 3 into a single measure using the same maximum value for classification as above.

Measures of Potential Explanatory Variables

Three aspects of gay-related stress were assessed at Time 1. The experience of gay-related stressful life events (e.g., arguments with others, losing a close friend, or being physically assaulted due to one’s homosexuality/bisexuality) was assessed using a 12-item checklist of events experienced in the past 3 months (see Rosario, Schrimshaw, Hunter & Gwadz, 2002 for the items and prevalence). The count of stressful events was positively skewed; thus we computed a dichotomous measure indicating whether female youths had experienced no events (0) or one or more events (1). In addition, we assessed two aspects of internalized gay-related stress, that is, internalized homophobia. A 33-item modified version of the Nungesser Homosexual Attitudes Inventory (NHAI; Nungesser, 1983) was administered using a 4-point response scale from “disagree strongly” (1) to “agree strongly” (4). The original NHAI was modified for LGB youths by simplifying the language and generalizing the item content to include female youths. The NHAI was selected because more contemporary measures (e.g., Shidlo, 1994) had not been developed at the start of this study. A factor analysis of this measure on the full sample of youths in this study identified 2 factors: negative attitudes toward homosexuality, with 11 items (e.g., “My [homosexuality/bisexuality] does not make me unhappy”), and discomfort with others knowing about my homosexuality/bisexuality, with 12 items (e.g., “If my straight friends knew of my [homosexuality/bisexuality], I would feel uncomfortable”). The mean of each factor was computed (α_attitudes = .75 and α_discomfort = .89 for the female youths). The attitudes toward homosexuality measure was negatively skewed, therefore we used the exponential e to normalize the distribution.

Three aspects of emotional distress were assessed at Time 1. Depressive and anxious symptoms during the past week were measured using the Brief Symptom Inventory (BSI; Derogatis, 1993), with its 5-point response scale ranging from “not at all” (0) to “extremely” (4) distressing. The mean of the 6 depressive symptoms items and the mean of the 6 anxious symptoms items were computed, with higher scores indicating greater distress (α_depression = .82 and α_anxiety = .79 for the female youths). The BSI was selected because it has been validated among adolescent samples (Derogatis, 1993) and it or its parent scale, the Symptom Checklist-90 (Derogatis, 1983), has been and continues to be used extensively with LGB youths (e.g., D’Augelli, 2002; Hershberger & D’Augelli, 1995; Rotheram-Borus et al., 1995). In addition, conduct problems were assessed using a 13-item index (e.g., skipping school, vandalism, stealing, fighting, running away). Items were based on conduct problems identified in the DSM-III-R (American Psychiatric Association, 1987). This measure has been previously used in a study of gay and bisexual male youths (Rotheram-Borus et al., 1995). A count of the number of problems was computed.

The Marlowe-Crowne Social Desirability Scale (Crowne & Marlowe, 1964) was self-administered at Time 1 using a true-false response scale to assess the degree to which youths provided socially desirable responses. Two of the 33 items were removed because they were inappropriate for youths. A factor analysis revealed 12 items that loaded on a single factor. A count of these 12 items was computed as a measure of socially desirable response bias (α = .77 for the female youths).

Results

Bivariate Relations

The mean differences between butches and femmes on substance use and abuse are presented in Table 1. All significant and trend differences indicated that butches more frequently used and consumed larger quantities of tobacco, alcohol, and marijuana over time than femmes. Butches also reported more symptoms of substance abuse at Time 3.

Table 1.

Butch / Femme Differences in Substance Use and Abuse: Mean Comparisons

	Butch (n = 33)	Femme (n = 39)	Statistical Test	Effect Size
	M (SD)	M (SD)	F	r
Cigarette Use: Quantity
Time 1	2.0 (1.9)	1.0 (1.4)	6.89*	.30
Time 2	1.8 (1.7)	0.9 (1.3)	4.51*	.26
Time 3	1.8 (1.7)	0.9 (1.2)	5.27*	.30
Alcohol Use: Frequency
Time 1	1.4 (0.8)	0.8 (0.7)	12.39**	.39
Time 2	1.8 (1.0)	1.4 (1.1)	1.96	.18
Time 3	1.7 (0.9)	1.6 (0.9)	0.30	.07
Alcohol Use: Quantity
Time 1	1.9 (1.1)	1.3 (1.2)	4.56*	.25
Time 2	2.0 (1.1)	1.5 (1.2)	2.46	.20
Time 3	2.0 (0.9)	1.6 (1.0)	2.53	.21
Marijuana Use: Frequency
Time 1	0.8 (0.8)	0.4 (0.6)	3.44+	.22
Time 2	1.1 (0.9)	0.6 (0.7)	5.51*	.29
Time 3	1.0 (0.9)	0.6 (0.8)	3.76+	.24
Marijuana Use: Quantity
Time 1	0.8 (0.9)	0.5 (0.7)	3.45+	.22
Time 2	1.5 (1.6)	0.7 (1.0)	4.86*	.27
Time 3	0.9 (0.8)	0.7 (0.8)	1.36	.15
Substance Abuse Symptoms
Time 1	1.7 (0.7)	1.5 (0.9)	0.72	.10
Time 2	1.5 (0.6)	1.4 (0.5)	0.40	.09
Time 3	1.7 (0.8)	1.4 (0.6)	4.32*	.26

^**p < .01

^*p < .05

⁺p < .07

Several potential confounds (e.g., age, race/ethnicity, SES, sexual identity, and social desirability) were examined. Butches were significantly older than femmes (r = .25). Butches were more likely to identify as lesbian (91%), whereas femmes were more likely to identify as bisexual (54%), χ² (1, N = 71) = 15.54, p < .001. Although no significant relations were found between social desirability and butch/femme identity, we thought it important to account for any potential social desirability bias in subsequent analyses.

Multivariate Relations

To investigate whether the observed butch/femme differences in substance use and abuse were due to the possible confounds, linear regression analyses were conducted controlling for age, lesbian/bisexual identity, and social desirability. Generally, the inclusion of these covariates had no effect on the original magnitude of bivariate relations (see Table 2). Specifically, young butches still were found to report significantly more cigarette use, more frequent alcohol use (at Time 1 only) and more frequent marijuana use than did young femme women. In some instances the inclusion of the covariates enhanced the magnitude of the bivariate relations. Whereas the bivariate findings found butches to consume a greater quantity of alcohol only at Time 1, with the inclusion of the confounds, young butches were found to report drinking more alcohol than femmes at all three assessments. Only infrequently did the covariates explain (i.e., eliminate) the bivariate relations. Specifically, although bivariate analyses found that butches reported both greater quantity of marijuana use at Time 2 and more substance abuse at Time 3 than young femmes, after imposing controls these two associations became nonsignificant.

Table 2.

Butch / Femme Differences in Substance Use and Abuse: Multivariate Analyses

	Demographic Controls1		Mediational Analysis2
	Butch (1) vs. Femme (0)		Butch (1) vs. Femme (0)
	β	SE	β	SE
Cigarette Use: Quantity
Time 1	.33*	.13	.14	.12
Time 2	.35*	.14	.20	.14
Time 3	.38*	.15	.24	.15
Alcohol Use: Frequency
Time 1	.43**	.13	.40**	.13
Time 2	.15	.15	na
Time 3	.20	.15	na
Alcohol Use: Quantity
Time 1	.28*	.14	.28+	.14
Time 2	.27+	.14	.24	.15
Time 3	.29*	.14	.31+	.16
Marijuana Use: Frequency
Time 1	.26+	.14	.18	.15
Time 2	.32*	.14	.27+	.15
Time 3	.29+	.14	.22	.14
Marijuana Use: Quantity
Time 1	.24+	.14	.17	.15
Time 2	.19	.14	na
Time 3	.14	.15	na
Substance Abuse Symptoms
Time 1	.05	.14	na
Time 2	.04	.16	na
Time 3	.19	.14	na

Note: na = not applicable because butch/femme was not associated with this outcome; thus, there was no significant relation to explain.

¹Controlling for age, sexual identity (lesbian vs. bisexual), and social desirability at Time 1.

²Controlling for age, sexual identity, social desirability, depressive symptoms, anxious symptoms, conduct problems, gay-related stressful life events, and internalized homophobia (i.e., attitudes toward homosexuality and comfort with others knowing about their sexuality) at Time 1.

^**p < .01

^*p < .05

⁺p < .09

Potential Theoretical Explanations

In order to explore potential reasons why young butches were found to report greater substance use than young femmes, another set of regression analyses was conducted in which we controlled for both the confounds and potential Time 1 explanatory factors of gay-related stressful events, internalized homophobia (i.e., negative attitudes toward homosexuality/ bisexuality and discomfort with other’s knowing about one’s homosexuality/bisexuality), and emotional distress (symptoms of depression, anxiety, and conduct problems). As the findings in the last two columns of Table 2 indicate, these theoretical factors accounted for the relations between butch/femme and quantity of cigarette use at all time periods. Gay-related stress, internalized homophobia, and emotional distress also attenuated the relations between butch/femme and frequency and quantity of marijuana. However, these theoretical factors had little impact on the relations between butch/femme and frequency or quantity of alcohol use.

Discussion

Although lesbian and bisexual female youths have elevated rates of substance use relative to heterosexual peers (Eisenberg & Wechsler, 2003; Russell et al., 2002; Ziyadeh et al., 2007), little research has examined the potential reasons for these disparities. The current report suggests that gender atypicality may be one critical factor in understanding which lesbian and bisexual women are at greatest risk for substance use. Specifically, young women with a more butch self-presentation were found to use alcohol and marijuana more frequently and to consume larger quantities of cigarettes and alcohol than young femme women. This pattern of findings generally held even after controlling for potential confounders of these relations, namely age, sexual identity as lesbian or bisexual, and social desirability.

We examined potential theoretical reasons why young butch women reported using more alcohol, cigarettes, and marijuana. Specifically, aspects of emotional distress and gay-related stress (both internalized homophobia and external stressful events) were found to account for the higher levels of cigarette and marijuana use found among young butch women. This suggests that tobacco and marijuana may be used by young butch women as a coping strategy to manage stress or self-medicate emotional distress potentially caused by stress. As noted earlier, butches may experience more stressful experiences and consequent distress than femmes because they are more easily identified as lesbian and, thus, are more likely to experience anti-gay abuse, prejudice, discrimination, and violence. In contrast, the greater frequency and quantity of alcohol use found among young butch women were largely unexplained by gay-related stress and emotional distress. Thus, alcohol use by young butch women may not be a coping strategy and other factors (e.g., sensation seeking, more masculine role norms regarding alcohol consumption) may explain the greater alcohol use found among young butches relative to femmes.

The findings reported here have a number of important research and intervention implications. First, they highlight the need to consider gender atypicality as a potential risk factor for substance use among young lesbian and bisexual women. Although not all young lesbian and bisexual women were found to be at risk for high levels of substance use, young butch women were at particular risk. Consequently, intervention efforts to reduce substance use may want to target young butch women. Further, the findings suggest that interventions to reduce substance use among lesbian and bisexual women may need to address the experiences of gay-related stress and emotional distress, as these factors were found to account for the higher levels of cigarette and marijuana use found among the young butch women. Although practical considerations may limit the feasibility of targeting only young butch women for intervention, substance use prevention and treatment efforts with all young lesbian and bisexual women may benefit from addressing the issues of gay-related stress and emotional distress.

Despite the possibility that the current research is limited by the small sample of young urban lesbian and bisexual women, the report’s findings provide a valuable preliminary examination of the role of butch/femme self-presentation in understanding the substance use of young lesbian and bisexual women, including the roles of gay-related stress and emotional distress in these relations.

Glossary

Butch

a woman (typically lesbian or bisexually identified) who identifies as having, or who is perceived to have, behaviors, appearance, or other characteristics considered masculine by society.

Femme

a woman (typically lesbian or bisexually identified) who identifies as having, or who is perceived to have, behaviors, appearance, or other characteristics considered feminine by society.

Emotional Distress

the extent to which individuals experience negative psychological symptoms including, for example, symptoms of depression and anxiety.

Gay-related Stress

the experience of stigmatization (or the internalization of this stigmatization) for being, or being perceived to be, gay, lesbian, or bisexual in a society in which homosexuality is negatively sanctioned.

Gender Atypical Self-Presentation

behaviors, appearance, or other characteristics that society considers more typical of the other sex.

Sexual Identity

one’s self-identification as lesbian, gay, bisexual, or heterosexual.

Biography

Margaret Rosario, Ph.D., is an Associate Professor of Psychology at the City University of New York – City College and Graduate Center. She received her Ph.D. in psychology from New York University. Her research interests include the relation of identity to health, the intersection of multiple identities, and the relation of violence to health.

Eric W. Schrimshaw, M.A., is a doctoral candidate in social/personality and health psychology at the City University of New York – Graduate Center. In addition, he is a project director at the Center for the Psychosocial Study of Health and Illness, at the Mailman School of Public Health, Columbia University. His research interests include the role of sexual identity, identity disclosure, and identity development on health and well-being.

Joyce Hunter, D.S.W., is a Research Scientist at the HIV Center for Clinical and Behavioral Studies, New York State Psychiatric Institute, and an Assistant Professor in the Departments of Psychiatry and Sociomedical Sciences at Columbia University. Dr. Hunter received her doctorate in social welfare (DSW) from The City University of New York - Graduate Center. Her research and clinical interests include the design and implementation of interventions for gay, lesbian, and bisexual adolescents.