Abstract
Objectives. This study measured the prevalence of battering victimization (i.e., experience of psychological/symbolic, physical, and sexual battering) among men who have sex with men (MSM) and identified characteristics of these men.
Methods. A probability-based sample of 2881 MSM living in 4 cities completed telephone interviews between 1996 and 1998.
Results. Prevalence estimates were 34% for psychological/symbolic battering, 22% for physical battering, and 5% for sexual battering. The strongest demographic correlate independently associated with all forms of battering was age 40 or younger, whereas education and HIV serostatus were associated with physical and psychological/symbolic violence.
Conclusions. Rates of battering victimization among urban MSM are substantially higher than among heterosexual men and possibly heterosexual women. Public health efforts directed toward addressing intimate partner battering among these men are needed.
Despite decades of research on interpersonal violence within heterosexual relationships, very little is known about intimate partner violence among same-gender partners, particularly among men who have sex with men (MSM). Published estimates of partner abuse among MSM range from 12% to 36%.1–3 Despite the fact that these estimates are roughly comparable to cited measures of domestic violence among heterosexual women4–6 (but slightly lower than those among lesbians7,8), intimate partner violence among MSM has been virtually ignored as a public health problem.
A number of factors contribute to this lack of information. First, national probability samples that measure violence among the general population do not ask about sexual orientation or same-gender battering. Second, constructing representative samples of MSM is very difficult and expensive, and studies with such samples generally have not measured partner violence. Third, gender-based ideologies about partner violence have had the unintended effect of repressing discourse on or study of same-gender battering and contributing to the myth that men are perpetrators rather than victims of partner violence.9,10 Fourth, the few studies of same-gender partner abuse have serious methodological flaws (nonrandom sampling procedures, small sample sizes, and poor research designs) that inflate prevalence estimates7 and prevent examination of risk factors of partner violence. Finally, these studies have failed to use standard definitions of battering or methodologically sound and sensitive procedures to index abuse.11 Our study significantly expands the current state of the knowledge on battering among MSM by addressing many of these limitations, most notably by using a large, probability-based sample of MSM, standard definitions of abuse,12,13 and sensitive and rigorous data collection procedures.
Because of the paucity of surveillance research, it is not known how the burden of battering among MSM compares with that of better-studied populations. What little is known suggests that these men share similar risk “profiles” with battered women: lower income, unemployment, family histories of violence, childhood sexual abuse, depression, and heavy substance use.7,8,14,15 Some authors have argued that unlike most battered women, battered MSM are more likely to experience partner violence within a social context of homophobia and HIV/AIDS.9,14 However, recent research indicates that HIV-infected women are significantly exposed to domestic violence4 and that it is the social inequalities related to economic marginalization, substance abuse, and childhood trauma that contribute to the occurrence of intimate partner abuse.16
To build effective intervention and prevention programs, it is first necessary to enumerate accurately the prevalence of same-gender battering among a representative sample of MSM and to identify the basic characteristics of battered men. In this study, we measured 3 types of partner violence: physical, sexual, and psychological/symbolic. We discuss implications for public health efforts and prevention and treatment research to reduce the scope of partner violence among MSM.
METHODS
Sample Construction
The sampling procedures have been described in detail elsewhere.17–19 Preliminary work drew on a number of data sources (e.g., MSM AIDS cases, gay mail commercial mailing lists, 1990 census data on male–male partnered households) to map where MSM resided in 4 urban centers (San Francisco, Los Angeles, New York, and Chicago). The sample frame included telephone exchanges overlying the selected zip codes; the estimated cost per interview was under $1000. Disproportionate and adaptive sampling techniques were used to construct a random-digit-dial sample for designated areas in each city. “Disproportionate sampling” means that we sampled more from exchanges in areas with greater densities of MSM and less from exchanges in areas with lower densities of MSM. “Adaptive sampling” means that as we learned which telephone exchanges yielded the best chance of finding households containing an MSM (“MSM eligible”), we used that information to target release of other exchanges later in the study. Both of these schemes were employed20 to maximize “hit rates” and minimize costs (which are substantial for a study of this type).
Of the 95 208 households screened, approximately 55 000 were eligible. We first screened for zip code (43 545), then for adult male (aged 18 years or older) (27 867), and finally for MSM eligibility (3700). The first available adult male informant was asked a series of questions about samegender sex and then asked similar questions for other men in the household. To reduce self-disclosure bias in the screening interview, we used only male interviewers, because men have been found to be more likely to disclose same-gender sexual behavior to male interviewers in methodological studies.21 We also instituted procedures that past experience had shown would make respondents feel more comfortable with questions of a sensitive nature (e.g., privacy and study credibility assurances). We selected all men who reported same-gender sexual behavior since age 14 years or who self-labeled as homosexual, gay, or bisexual. For households with multiple men eligible, 1 man only was randomly selected. The obtained proportion of MSM households within each zip code ranged from a low of 1.3% to a high of 30.8%. Sample weights were developed to reflect probability of selection, nonresponse, and noncoverage. The sample was also adjusted to maintain proportionality between cities on the basis of the estimated size of each city’s MSM population.17 Interviews were conducted between November 1996 and February 1998. Of 3700 eligible men, 2881 completed interviews (acceptance rate = 77.9%).
Battering Victimization
To obtain accurate and comparable estimates of battering victimization, a modified version of the Conflict Tactics Scale22 was used. The introduction was modified to be culturally specific to MSM. Respondents were asked to report “unwanted physical or emotional violence” from a boyfriend or same-gender partner during the past 5 years. A 5-year instead of a 1-year recall period was used in this first-ever surveillance of intimate partner abuse among MSM to capture a wider picture of battering experiences.
Three types of battering victimization were measured. Psychological/symbolic battering was defined as having experienced at least 1 of the following: being verbally threatened, demeaned in front of others, ridiculed for his appearance, forced to get high or drunk, or stalked; or having property destroyed or damaged. Physical battering was defined as being hit with fists or an open hand, hit with an object, pushed or shoved, or kicked; or having something thrown at him. Sexual battering was defined as having been forced to have sex. Two “global” measures of battering victimization were also constructed: any battering (experiencing at least 1 type of battering) and multiple battering (experiencing 2 or 3 types of battering).
Sociodemographic Characteristics
Standard sociodemographic factors were measured to identify the demographic distribution of battering victimization among MSM. Respondents were asked their age (by decade), educational attainment (high school, college, graduate degree, professional degree), race/ethnicity, employment status (full-time, part-time, other employment status), income (by $10 000 increments), self-defined sexual orientation (gay or homosexual, bisexual, heterosexual, other such as “don’t use labels”), HIV serostatus (HIV negative, HIV positive, never tested), and city of residence.
Analyses
The prevalence of all types of battering victimization was reported for the total sample (n = 2881). Bivariate analyses (χ2 tests and simple logistic regressions) of each characteristic and battering outcome were determined. Multivariate logistic regression procedures (simultaneous entry) were then conducted. We performed these procedures with SPSS 10 (SPSS Inc, Chicago, Ill). We calculated corrected χ2 tests and adjusted standard errors by using the SVYTAB and SVYLOGIT procedures in Stata (Stata Corp, College Station, Tex) to correct for the clustered nature of the sample as well as the weighting. Results reported here were from these final Stata runs.
RESULTS
Prevalence of Battering Victimization
During the previous 5 years, 34% (95% confidence interval [CI] = 31.8%, 36.2%) of the urban MSM in our study experienced psychological/symbolic abuse, 22.0% (95% CI = 20.1%, 24.0%) experienced physical abuse, and 5.1% (95% CI = 4.1%, 6.4%) experienced sexual abuse. Some type of battering victimization was reported by 39.2% (95% CI = 37.0%, 41.5%) of the respondents, with 18.2% (95% CI = 16.5%, 20.1%) reporting multiple battering (i.e., more than 1 type of battering during the previous 5 years).
Characteristics of Battering Victimization
Table 1 ▶ presents prevalence estimates of battering victimization among urban MSM stratified by selected demographic characteristics. Age and education were associated with all forms of battering (all P values < .02), and HIV serostatus was associated with all forms except for sexual battering. Univariate logistic regression results (odds ratios and 95% confidence intervals) revealed that MSM aged 40 years or younger were substantially more likely than MSM aged 60 years or older to experience sexual battering (for ages 18–29 years, odds ratio [OR] = 6.2, 95% CI = 1.4, 27.5; for ages 30–39 years, OR = 4.8, 95% CI = 1.1, 20.4). MSM with graduate or professional degrees were substantially less likely to experience any form of partner abuse than were MSM with a college degree or less. Compared with HIVnegative MSM, HIV-positive men were more likely to be victims of battering (all types except sexual), whereas MSM who had never been tested for HIV were less likely. Surprisingly, MSM who were employed part-time were 35% less likely than those employed full-time to experience any partner abuse (OR = 0.66, 95% CI = 0.46, 0.95). None of the battering outcomes were associated with race/ethnicity, income, sexual orientation, or city of residence.
TABLE 1.
—Selected Sociodemographics by Prevalence Estimates of Battering Victimization During Previous 5 Years Among Urban Men Who Have Sex With Men (N = 2881)
| Type of Battering Victimization | |||||
| Characteristic (n) | Psychological/Symbolic, % (95% CI) | Physical, % (95% CI) | Sexual, % (95% CI) | Multiple,a % (95% CI) | Any,b % (95% CI) |
| Age, y | |||||
| 18–29 (565) | 39.6 (34.3, 45.2)* | 25.3 (20.8, 30.4)* | 7.9 (5.1, 12.0)* | 21.6 (17.3, 26.5)* | 45.8 (40.3, 51.3)* |
| 30–39 (1122) | 38.8 (35.3, 42.4)* | 27.1 (23.8, 30.6)* | 6.1 (4.4, 8.5)* | 22.2 (19.1, 25.5)* | 45.6 (41.9, 49.3)* |
| 40–49 (718) | 33.6 (29.5, 38.0)* | 20.1 (16.8, 23.9)* | 2.8 (1.6, 5.0)* | 16.8 (13.7, 20.4)* | 37.6 (33.4, 42.1)* |
| 50–59 (287) | 17.2 (12.7, 22.8)* | 10.8 (7.2, 15.9)* | 3.8 (1.9, 7.4)* | 8.9 (5.6, 13.8)* | 20.6 (15.7, 26.5)* |
| ≥60 (184) | 14.8 (9.7, 21.9)* | 5.3 (2.7, 10.1)* | 1.4 (0.3, 5.3)* | 4.6 (2.3, 9.2)* | 16.0 (10.7, 23.2)* |
| Race/ethnicity | |||||
| African American (123) | 34.7 (25.6, 45.1) | 24.2 (16.1, 34.7) | 8.0 (3.8, 16.1) | 19.6 (12.5, 29.4) | 41.1 (31.4, 51.5) |
| White (2266) | 34.0 (31.5, 36.6) | 21.7 (19.7, 24.0) | 4.7 (3.6, 6.1) | 18.1 (16.1, 20.3) | 39.1 (36.6, 41.7) |
| Asian/Pacific Islander (120) | 27.8 (18.7, 39.3) | 14.7 (8.2, 24.9) | 2.6 (0.8, 8.5) | 12.6 (6.7, 22.7) | 31.1 (21.5, 42.7) |
| Latino (273) | 34.5 (27.5, 42.2) | 23.4 (17.4, 30.6) | 8.5 (4.6, 15.1) | 18.9 (13.3, 26.1) | 41.3 (34.0, 49.0) |
| Native American (77) | 40.5 (27.8, 54.5) | 28.8 (18.2, 42.5) | 3.9 (1.2, 12.2) | 23.5 (13.8, 37.1) | 46.2 (33.1, 60.0) |
| Education | |||||
| High school diploma or less (858) | 37.4 (33.3, 41.8)* | 24.8 (21.3, 28.7)* | 6.7 (4.7, 9.5)* | 20.9 (17.6, 24.6)* | 43.1 (38.8, 47.4)* |
| College degree (1298) | 34.2 (31.0, 37.6)* | 23.4 (20.6, 26.4)* | 5.4 (3.9, 7.6)* | 19.5 (16.8, 22.5)* | 40.0 (36.7, 43.4)* |
| Graduate/professional degree (724) | 29.5 (25.6, 33.7)* | 16.1 (13.0, 19.7)* | 2.6 (1.5, 4.5)* | 12.9 (10.1, 16.3)* | 33.5 (29.4, 37.8)* |
| HIV serostatus | |||||
| Positive (442) | 38.5 (33.0, 44.3)* | 28.7 (23.7, 34.2)* | 6.4 (3.6, 11.1) | 24.6 (19.9, 30.0)* | 43.1 (37.5, 49.0)* |
| Negative (1927) | 34.6 (32.0, 37.3)* | 21.4 (19.3, 23.8)* | 4.9 (3.7, 6.4) | 17.6 (15.5, 19.8)* | 40.1 (37.5, 42.9)* |
| Don’t know/never tested (292) | 20.6 (15.3, 27.1)* | 12.7 (9.0, 17.5)* | 3.5 (1.6, 7.5) | 9.7 (6.5, 14.1)* | 25.6 (19.8, 32.5)* |
| Employment status | |||||
| Full-time (1975) | 35.1 (32.6, 37.9) | 22.3 (20.2, 24.7) | 5.0 (3.8, 6.5) | 18.2 (16.2, 20.5) | 40.9 (38.2, 43.6)* |
| Part-time (219) | 29.0 (22.3, 36.7) | 17.4 (12.1, 24.4) | 4.3 (2.2, 8.5) | 16.9 (11.7, 23.9) | 31.3 (24.4, 39.1)* |
| Not employed (507) | 33.9 (31.7, 36.2) | 22.2 (18.0, 27.0) | 5.9 (3.6, 9.4) | 18.5 (14.6, 23.1) | 36.2 (31.2, 41.5)* |
| Income, $ | |||||
| < 20 000 (417) | 35.9 (30.4, 41.8) | 22.1 (17.7, 27.3) | 9.0 (6.0, 13.4) | 20.7 (16.3, 25.9) | 39.6 (33.9, 45.5) |
| 20 001–40 000 (694) | 37.3 (33.1, 41.7) | 22.2 (18.8, 26.0) | 4.3 (2.9, 6.4) | 20.4 (17.1, 24.2) | 40.7 (36.5, 45.1) |
| 40 001–60 000 (541) | 35.0 (30.1, 40.1) | 24.6 (20.3, 29.5) | 4.7 (2.6, 8.6) | 18.9 (14.9, 23.5) | 41.1 (36.0, 46.3) |
| 60 001–80 000 (341) | 27.4 (22.0, 33.6) | 20.2 (15.4, 26.1) | 2.9 (1.3, 6.0) | 15.2 (11.1, 20.5) | 34.0 (28.0, 40.5) |
| 80 001–100 000 (211) | 29.7 (22.7, 37.8) | 15.1 (10.2, 21.8) | 3.4 (1.3, 8.9) | 11.6 (7.5, 17.3) | 36.3 (28.4, 45.0) |
| > 100 000 (464) | 32.6 (27.1, 38.7) | 22.5 (17.7, 28.1) | 5.1 (2.7, 9.4) | 17.1 (12.8, 22.6) | 39.2 (37.0, 41.5) |
Note. CI = confidence interval. “n” values do not add to 2881 owing to missing data.
aTwo or more types of battering.
bAny type of battering.
*P < .05.
Independent associations of psychological/symbolic, physical, and sexual battering with demographic characteristics were identified via multivariate logistic regression procedures (Table 2 ▶). These findings were similar to those for “global” measures of battering victimization. The strongest demographic correlate of partner violence—any or multiple forms—was age. Compared with the odds for MSM aged 60 years or older, the odds of experiencing any battering were 3.8 (95% CI = 2.1, 6.7) for 18- to 29-year-old MSM, 3.9 (95% CI = 2.3, 6.7) for 30- to 39-year-old MSM, and 2.7 (95% CI = 1.6, 4.7) for 40- to 49-year-old MSM. Similarly, MSM younger than 40 were about 6 times as likely to report multiple forms of partner violence as were MSM aged 60 or older, whereas 40- to 50-year-old men were about 4 times as likely. HIV-positive MSM were 11/2 times as likely to experience multiple battering as were HIV-negative MSM (OR = 1.5, 95% CI = 1.1, 2.2); however, such differences were not found for any abuse. MSM who had never been tested for HIV were less likely than HIV-negative MSM to experience multiple battering (OR = 0.56, 95% CI = 0.35, 0.91) or any battering (OR = 0.60, 95% CI = 0.42, 0.86).
TABLE 2.
—Independent Characteristics of Battering Victimization During Previous 5 Years Among Urban Men Who Have Sex With Men (N = 2594)
| Type of Battering Victimization | |||
| Characteristic | Psychological/Symbolic, OR (95% CI) | Physical, OR (95% CI) | Sexual, OR (95% CI) |
| Age, y | |||
| 18–29 | 3.2 (1.8, 5.8)* | 6.1 (2.6, 14.0)* | 5.3 (1.1, 26.2)* |
| 30–39 | 3.3 (1.9, 5.7)* | 6.8 (3.1, 15.2)* | 5.2 (1.1, 25.0)* |
| 40–49 | 2.5 (1.4, 4.5)* | 4.4 (1.9, 9.9)* | 2.0 (0.39, 10.0)* |
| 50–59 | 1.0 (0.54, 1.9)* | 2.1 (0.84, 5.1)* | 3.1 (0.58, 16.2)* |
| ≥60 | 1.0 | 1.0 | 1.0 |
| HIV serostatus | |||
| Negative | 1.0 | 1.0 | NS |
| Positive | 1.2 (0.87, 1.6)* | 1.5 (1.1, 2.1)* | NS |
| Never tested | 0.55 (0.37, 0.80)* | 0.63 (0.41, 0.97)* | NS |
| Education | |||
| High school | NS | 1.1 (0.80, 1.4)* | NS |
| College | NS | 1.0 | NS |
| Graduate/professional | NS | 0.67 (0.49, 0.92)* | NS |
Note. OR = odds ratio; CI = confidence interval; NS = not significant in final model.
*P for given odds ratio < .03.
DISCUSSION
The prevalence of battering within the context of intimate partner relationships was very high among this probability-based sample of urban MSM. Approximately 2 of 5 MSM (39%) reported experiencing at least 1 type of battering by a partner during the previous 5 years, with almost 1 of 5 (18%) experiencing multiple forms of battering (34% reported psychological/symbolic violence, 22% physical violence, and 5% sexual violence). In a nationally representative sample1 of heterosexual men (defined in the study as men who reported cohabitation with women), 7.7% reported lifetime physical or sexual partner violence, compared with 23% (95% CI = 21.5%, 25.4%) of our urban MSM who reported such battering during the previous 5 years. Because lifetime rates are generally higher than rates for a 5-year recall period, it is likely that a substantially greater number of MSM than of heterosexual men have experienced lifetime victimization. Similarly, Zierler and colleagues16 found that among a nationally representative sample of HIV-infected individuals, 7.5% of heterosexual men had experienced some type of battering (since HIV diagnosis), compared with 39% of our sample (within the past 5 years).
To examine how the burden of partner violence among our sample of MSM compares with that among a representative sample of heterosexual females, we located 2 studies conducted during the past decade.23,24 Precise comparisons are not feasible because different items from the Conflict Tactics scale22 were used, as well as different recall periods. The 2 national studies of heterosexual women defined partner violence as “severe” (i.e., being hit, kicked, bitten, choked, beaten up, threatened with a knife or gun, or injured by a knife or gun) or “total” (i.e., all previous items plus being shoved, pushed, grabbed, slapped, spanked, or having something thrown at them). Neither of these definitions was an exact comparison to our battering outcomes. Whereas these studies measured partner abuse during the past year, our study indexed partner abuse during the past 5 years. Furthermore, important demographic factors (e.g., age, education, socioeconomic status) could not be controlled. Nonetheless, rates of physical battering can be placed side by side to suggest how the burden of intimate partner abuse between MSM and women might compare.
In general, the 5-year prevalence of physical battering among urban MSM (22.0%) was significantly higher than either the annual prevalence of severe violence (3.4%) or the annual prevalence of total violence (11.6%) among a representative sample of women who were married or cohabiting with men.24 Whereas the lifetime estimates for severe domestic violence among these women24 were still below our 5-year estimate among urban MSM, rates of lifetime total domestic violence among women were higher (30%) than our 5-year rate among MSM.
Our estimates are substantially higher than those reported for heterosexual men1,16 and higher than or comparable to those reported for heterosexual women.23–26 This study demonstrates that intimate partner abuse among urban MSM is a very serious public health problem. It sheds light on a subject that has long been taboo both within and outside this MSM community—that is, men are also victims of battering and not solely perpetrators.
Basic demographic factors independently associated with battering victimization were age, HIV serostatus, and education. Younger age was the strongest and most consistent demographic correlate of all forms of battering in this study. This finding has been supported among representative samples of HIVinfected MSM16 and heterosexual women.27,28 It is possible that battering victimization diminishes with older age for a number of reasons. It may be that older persons are more likely to have external and internal resources for protection than are younger MSM, who may be more vulnerable or easily influenced or may have fewer options to remove themselves from the battering situation. Or it may be that if MSM tend to choose similarly aged partners, battering could decrease across the years because violence perpetration tends to decrease with age, possibly as a result of hormonal changes among aging perpetrators of battering.
HIV serostatus was associated with all forms of battering except sexual violence. Unexpectedly, MSM who had never been tested for HIV were substantially less likely to report partner violence than were MSM who knew that they were HIV negative; however, how this factor is protective is not clear from these limited data. Consistent with previous findings,16,29 HIV-infected MSM were more vulnerable to physical battering and to multiple forms of partner victimization, but what role HIV infection plays within the context of victimization was not assessed. And whereas Zierler and colleagues found battering to be disproportionately prevalent among Latino or unemployed MSM,16 such findings did not emerge in our study.
Limitations
Our measurement of battering victimization did not use standard items or a standard recall period; therefore, comparability with national data sets is limited. Despite sampling improvements, prevalence estimates of battering among MSM may still be underreported. It is likely that battering is higher among disenfranchised MSM such as those who are without telephones or who are marginally housed or homeless, subsegments of the MSM population that were undersampled by the procedures that we used. It is also unclear how our data compare with those for MSM across the United States. Analytic findings are correlational only and are narrowly focused on sociodemographic characteristics. These findings do not shed light on the contexts (personal, social, situational, or cultural) or dynamics of battering, nor do they reveal any information about the batterer or the severity or frequency of partner violence. Instead, these profiles help only to understand and to identify who is at risk, and to provide some hypotheses about the demographic distribution of this public health problem.
Implications and Future Directions
Because judicial, legislative, and public health systems do not recognize or are not aware of intimate partner abuse among MSM, serious social and structural changes are needed. To respond to this very serious public health problem, we need to develop and support shelters for battered MSM, educate and train law enforcement personnel about battering among MSM and how to respond to it, and expand preventive and clinical care7–10,14,29 for these men. A full range of medical and domestic violence services for MSM, particularly services targeting MSM aged 40 years or younger, are needed. Health professionals need to be able to appropriately screen, treat, and screen, treat, or recommend services for intimate partner abuse. Our society needs to understand that men are victims as well as perpetrators of violence.9,10 Equally intensive and multilayered public health efforts are also needed to intervene with and serve the perpetrators of violence among MSM. Surveillance, prevention, and intervention research on intimate partner abuse among MSM has not been well conducted. Sorely needed are theory-driven, longitudinal, mixed methodological and well-controlled studies that systematically elucidate the etiology, maintenance, context, and trajectories of partner violence among MSM.29 These studies could also help to identify how intimate partner abuse among MSM is similar to, and different from, such abuse among lesbians and heterosexual women. Finally, “best practices” research could be conducted to identify which treatment approaches work best to reduce the burden of same-gender battering among these men.
Acknowledgments
National Institute of Mental Health grant MH54320 provided primary support for this study.
This study would not have been possible without the extensive cooperation of the men who were willing to serve as project participants and the dedication of interviewers and project staff at Survey Methods Group of San Francisco. The random-digit-dial sample frame was constructed by Johnny Blair and Tim Tripplett at the Survey Research Center of the University of Maryland, in collaboration with Dr Graham Kalton at Westat. Drs Diane Binson, Tom Mills, Judith Moskowitz, Jay Paul, and Ron Stall were key contributors to the design, implementation, and success of this study.
Human Participant Protection No protocol approval was needed for this study.
G. L. Greenwood led the writing of the manuscript; he participated in the analysis plan, wrote all drafts, and revised the final manuscript. M. V. Relf contributed to the literature review, was involved in the conception of the original study question, participated in the analysis plan, and reviewed and edited drafts. B. Huang and J. A. Canchola conducted data analysis, assisted with interpretation, and participated in manuscript preparation. L. M. Pollack helped design and conduct the study, led data management, and participated in data analysis and interpretation and manuscript preparation. J. A. Catania, the principal investigator, designed the study; he was involved in the conception of the original study question and participated in manuscript preparation.
Peer Reviewed
References
- 1.Tjaden P, Thoennes N, Allison CJ. Comparing violence over the life span in samples of same-sex and opposite-sex cohabitants. Violence Vict. 1999;14:413–425. [PubMed] [Google Scholar]
- 2.Waldner-Haugrud LK, Vaden Gratch L, Magruder B. Victimization and perpetration rates of violence in gay and lesbian relationships: gender issues explored. Violence Vict. 1997;12:173–184. [PubMed] [Google Scholar]
- 3.Waterman CK, Dawson LJ, Bologna MJ. Sexual coercion in gay male and lesbian relationships: predictors and implications for support services. J Sex Res. 1989;26:118–124. [Google Scholar]
- 4.Cohen M, Deamant C, Barkan S, et al. Domestic violence and childhood sexual abuse in HIV-infected women and women at risk for HIV. Am J Public Health. 2000;90:560–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Coker AL, Smith PH, McKeown RE, King MJ. Frequency and correlates of intimate partner violence by type: physical, sexual and psychological battering. Am J Public Health. 2000;90:553–559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jones AS, Gielen AC, Campbell JC, et al. Annual and lifetime prevalence of partner abuse in a sample of female HMO enrollees. Womens Health Issues. 1999;9:295–305. [Google Scholar]
- 7.Burke LK, Follingstad DR. Violence in lesbian and gay relationships: theory, prevalence and correlational factors. Clin Psychol Rev. 1999;19:487–512. [DOI] [PubMed] [Google Scholar]
- 8.West CM. Leaving a second closet: outing partner violence in same-sex couples. In: Jasinski JL, Williams LM, eds. Partner Violence: A Comprehensive Review of 20 Years of Research. Thousand Oaks, Calif: Sage Publications; 1998:163–184.
- 9.Letellier P. Gay and bisexual male domestic violence victimization: challenges to feminist theory and responses to violence. Violence Vict. 1994;9:95–106. [PubMed] [Google Scholar]
- 10.Merrill GS. Understanding domestic violence among gay and bisexual men. In: Berger RK, ed. Issues in Domestic Violence. Thousand Oaks, Calif: Sage Publications; 1998:129–141.
- 11.Island D, Letellier P. Men Who Beat the Men Who Love Them. Binghamton, NY: Harrington Park Press Inc; 1991.
- 12.Council on Ethical and Judicial Affairs, American Medical Association. Physicians and domestic violence: ethical considerations. JAMA. 1992;267:3190–3193. [PubMed] [Google Scholar]
- 13.Sonkin DJ, Martin D, Walker L. The Male Batterer: A Treatment Approach. New York, NY: Springer; 1985.
- 14.Cruz JM, Firestone JM. Exploring violence and abuse in gay male relationships. Violence Vict. 1998;13:159–173. [PubMed] [Google Scholar]
- 15.Stermac L, Sheridan PM, Davidson A, Dunn S. Sexual assault of adult males. J Interpersonal Violence. 1996;11:52–64. [Google Scholar]
- 16.Zierler S, Cunningham WE, Anderson R, et al. Violence victimization after HIV infection in a US probability sample of adult patients in primary care. Am J Public Health. 2000;90:208–215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Binson D, Moskowitz J, Mills T, et al. Sampling men who have sex with men: strategies for a telephone survey in urban areas in the United States. In: Proceedings of the Section on Survey Research Methods, American Statistical Association. Alexandria, Va: American Statistical Association; 1996:68–72.
- 18.Blair J. A probability sample of gay urban males: the use of two-phase adaptive sampling. J Sex Res. 1999;36:39–44. [Google Scholar]
- 19.Catania JA, Osmond D, Stall RD, et al. The continuing HIV epidemic among men who have sex with men. Am J Public Health. 91:907–914. [DOI] [PMC free article] [PubMed]
- 20.Freeman H, Montgomery K, Lewis C, Corey C. Use of telephone surveys in AIDS-related community research. In: Fowler FJ, ed. Health Survey Research Methods. Rockville, Md: National Center for Health Services Research; 1989:173–179.
- 21.Catania JA, Binson D, Canchola JA, Pollack L, Hauk W, Coates T. Effects of interviewer gender, interviewer choice, and item wording on responses to questions concerning sexual behavior. Public Opin Q. 1996;60:345–375. [Google Scholar]
- 22.Straus MA. Measuring intrafamily conflict and violence: the Conflict Tactics (CT) Scale. J Marriage Fam. 1979;41:75–88. [Google Scholar]
- 23.Plichta SB, Weisman CS. Spouse or partner abuse, use of health services and unmet need for medical care in US women. J Womens Health. 1995;4:45–54. [Google Scholar]
- 24.Straus MA, Gelles RJ. Physical Violence in American Families. New Brunswick, NJ: Transaction Publishers; 1990.
- 25.Wilt S, Olson S. Prevalence of domestic violence in the United States. J Am Med Womens Assoc. 1996;3:77–82. [PubMed] [Google Scholar]
- 26.Straus MA, Gelles RJ, Stenmetz SK. Behind closed doors: violence in the American family. New York, NY: Anchor Books; 1980.
- 27.McCauley J, Kern DE, Kolodner K, et al. The “battering syndrome”: prevalence and clinical characteristics of domestic violence in primary care internal medicine practices. Ann Intern Med. 1995;123:737–746. [DOI] [PubMed] [Google Scholar]
- 28.Flitcraft A. From public health to personal health: violence against women across the life span. Ann Intern Med. 1995;123:800–801. [DOI] [PubMed] [Google Scholar]
- 29.Merrill GS, Wolfe VA. Battered gay men: an exploration of abuse, help seeking, and why they stay. J Homosex. 2000;39(2):1–30. [DOI] [PubMed] [Google Scholar]