Study On Gay Men's Living Conditions In Changsha

BackgroundThe homosexual phenomenon has been being gained much more attentions over the past 30 years. In the 1980s, human immunodeficiency virus (HIV) was found, and gay men was identified as high-risk group of AIDS. Since then, more efforts were spent in the field of gay men. Research on the influence of homosexual behaviors to themselves and our society has become hot issues in social science and public health. The relevant research subjects are as follows:the cause of homosexuality, lifestyles and living conditions of homosexuals, intimate relationship and violence, their psychosomatic health, homosexual behavsiors and AIDS/STDs, attitude towards homosexual, and so on.However, there are two major methodology problems in those research:①the representation of samples. Gay man was a hard-to-reach population. At present, there are two basic sampling strategies:one is using personal contacts of the researchers or other participants, which is usually described as "friendship networks" or "snowballing" procedures; The other is to recruite samples through organizations, newsletters, businesses, or social events of lesbian, gay, or bisexual communities. The samples selected through these two strategies are lack of diversity and representativeness. Currently, there is a new sampling method, Respondent-driven Sampling (RDS), in some relevant researches among hard-to-reach population. The RDS is the improvement of "snowballing", and similar to the probability sampling. However, it still has many limitations, and needs to be further applied and improved.②Most of studies on homosexual had no control group. Usually, the descriptions and analyses of a single group sample could not explain the phenomenon very deeply.Under the current circumstance of Chinese social culture, the characteristics and problems of the homosexual group's living conditions are:①Homosexual relationship is unstable, and their sex partners is changeful. There is no doubt that they are more vulnerable to be infected by various STDs (include AIDS);②Most lesbians and gay men have moral guilty and psychological stress. Once they meet some emotional problems, they may lose control and conduct suicide, self-mutilation, and so on;③Lesbians and gay men may bring themselves and their family members heavy mental distress. Further more, it is miserable if they get married with a heterosexual people. As an objective subculture, the homosexual population is in great need of social support, but the truth is that their living conditions are getting much worse than before by the exclusion of major society.This study was a cross-sectional research to outline the gay men's living conditions in Changsha. We measured some important issues, such as:socio-demographic features, dating relationships, sexual behavior features, suicide behaviors, life events, social support, dating-violence and life quality. Then, these datas were compared with a heterosexual control group. Results from this study could provide a scientific basis to improve the gay men's physical and psychological health and give evidence to take measures to combat the spread of AIDS and STDs.Goals and ObjectivesThe goal of the study is to learn about the urban gay men's living conditions, including livelihood, occupations, health status, social communication, etc., analyze the related factors of gay men's PCS (Physical Component Scale) and MCS (Mental Component Scale),and then to provide a scientific basis to improve the gay men's physical and psychological health and give evidence to take measures to combat the spread of AIDS and STDs. The specific objectives include:①To summarize the gay populations'socio-demographic features;②To describe the relationship between their sexul behaviors and STDs;③To explore the suicide ideation and behaviors, dating violence, dating relationship, negative life events, social support, and quality of life.④To analyze the related factors of gay men's PCS and MCS.⑤To describe the public's attitudes toward lesbians and gay men, and analyze the main related socio-demographic factors.⑥To translate and revise the Attitudes toward Lesbians and Gay Men Scale(ATLG, Gregory M. Herek,1994) to Chinese, and examine its reliability and validity under China's social-culture background.MethodsIn this study, we identified all respondents'sexual orientation according to their self-reports. Before the formal investigation, all respondents were asked to report their sexual orientation (homosexual, bisexual, heterosexual or transsexual). If the answer was not homosexual,then the investigation would be stopped immediately.In early stage, researchers had joined in many gay men parties and investigated 105 gay men. The results showed that 96.2% people usually used homosexual forum and QQ to interact with others. So we decided to sample gay men on internet randomly. In mainland, there are 5 famous gay websites:www.pybk.com, www.boysky.com, www.bf99.com, www.94gay.com and www.69py.com. Up to June,2009, the number of gay men in Changsha who registered on the 5 websites was nearly 12,000 person times, and, there were about 100 gay QQ groups, nearly 6,000 person-times. Therefore, the sampling frame included about 18,000 person-times. Then as considering the criterion (250-400) especially for STDs high risk group which recommended by the World Health Organization, we finally decided to investigate 500 samples.All subjects in control group were come from Physical Examination Center of XiangYa Hospital. In order to improve the balance of comparability, we selected the control subjects matached gender, educational background and age (±1) to the gay men. If they are college students, then the grade should be matached. With the permition of the sample, face-to-face interviews were conducted in all respondents.The content of investigation includes:①Social demography characteristics (with self-designed questionnaires);②characteristics of sexual behaviors and status of STDs (with self-designed questionnaires);③dating relationships and desire of heterosexual marriage (with self-designed questionnaires);④suicide ideation and behaviors (with Ronald C. Kessler's suicide questionnaire);⑤dating violence (with the Massachusetts Youth Risk Behavior Survey);⑥negative life event (Gay Affect and Life Events Scale, GALES);⑦social support (Social Support Rating Scale, SSRS);⑧quality of life (the version 2 of the SF-36 health survey);⑨public attitudes toward lesbians and gay men (ATLG, Gregory M. Herek,1994).Data Analysis:Established the database with EpiData 3.0, and then the data were processed with SPSS 17.0. A P-value of less than 0.05 was considered statistically significant. Bivariate analyses included chi-square tests. The mean scores on the different groups were compared with each other using the T square test, Analysis of Variance, Mann-Whitney Test or Kruskal-Wallis Test. Multiple Ordinal regression was used to examine the influence of PCS, MCS and public attitudes toward lesbians and gay men.Results:The study was conducted from March 5 to December 31,2009. Sixty gay subjects were unable to be contacted because their email, QQ, or phone numbers were all out of service, and 11 gay men rejected to be interviewed. The final number of the investigations is 429, and the response rate was 85.8%. Moreover,11 invalid questionnaires with incomplete answers or confused logic were eliminated. So, finally we got 418 valid gay samples. Addtionally, from November 1st to December 31st of 2009. we collected 330 straight men samples.1. Basic featuresIn total, the present study used a convenience sample consisting of 418 subjects whose age ranged from 16 to 44 years old (median age= 24, P25=21.8, P75=26.0); Han Chinese accounted for 94.5%; ethnic minorities accounted for 5.5%; 72.0% came from cites and 28.0% were rural population; the only child accounted for 57.9%; 91.4% of the participants were single,7.9% were married, and 3(0.7%) were divorced; 159(38.0%) were students; 39(9.3%) people's education level was less than college, 111(26.6%) were junior college degree,207(49.5%) were bachelor degree, 60(14.4%) were master degree or above, and 1 person did not report.Fifty eight point nine percent(246/418) had a job. Fifty three point eight percent(225/418) of respondents'payment was less than￥2000/month,25.6%(107/418) were between￥2000 and￥4000/month,15.3%(64/418) were￥4001-6000/month,4.8% (20/418) were￥6001-8000/month, and 2 (0.5%) persons were more than￥8000/month.2. Sex role in homosexul relationsAccording to the practical situations, the study divided the sex role in homo-sex relations in 10 sections (from 0 to 1). "Zero" means never have penetrative sex, and on the contrary, "1" means only have penetrative sex. The respondents were asked to make a choice according to their true situations. In this study,0,0.1,0.2,0.3 were defined passive group, which included 125 subjects (29.9%); 0.4,0.5,0.6 were defined moderate group, which included 137 respondents (32.8%); and 0.7,0.8, 0.9,1 were defined active group which included 156 (37.3%)subjects.3. Coming-OutComing-Out means that LGBT (lesbians, gay, bisexual, transsexual) discloses their sexual orientation in their life and work. In this study, 57.3% had ever come out. In those persons,5.0%(12/241) came out to their fathers,7.9%(19/241) came out to mothers,12.9%(31/241) once came out to their brothers and/or sisters,63.1%(152/241) came out to classmates,and 49.4%(119/241) came out to workmates or friends. It is noteworthy that none of 33 married gay men disclosed the sexual orientation to their wives.4. Dating relationships, desire of heterosexual marriage and procreationCompared to the control group, fewer gay man had stable love relationship during the research (χ2=21.671, P=0.001); their love duration was relatively short (x2=57.391, P=0.000), and 44.0% was less than 7 months. Among 385 unmarried gay men,13.5%(52/385) expressed definitely that they would not get married with women, and 37.1%(155/418) expected to marry a lesbian; additionally,74.9% (313/418) of gay men were eager for bearing or raising a child. 5. Characteritics of gay men's sexual behaviors(1) Looking for sex partner by internet. Compared with the control group, gay men were even more likely to look for sex partners by internet (χ2=322.883, P=0.000), and 85.6% of subjects reported they ever looked for a sex partner in internet. Obviously, over the last six months, the number of sexual partners of subjects who often look for sex partners in internet was more than others (Z=-4.220, P=0.000)(2) First sexual experience:among 418 gay men,6 men never had sex behaviors; as to others, the age of first sex ranged from 11 to 30 years old (median age=19, QR=4); and there were no significant difference between gay men and straight men; For the experience of first sex partner, 344 (82.3%) gay men's were male, and other 68 (16.3%) were female; in control group, there were 119 man never had sex behaviors,7(2.1%) subjects'first sex partners were male, and other 198 (60.0%) straight men's were female; and there was statistical difference between the two groups (χ2=485.122, P=0.000)(3) Sexual contact with female. There were 126(30.1%) gay men ever had sex with female, and in the last six months,9.8%(40/418) had sexual contacts with female.(4) Regular sexual partners. In the last 6 months,90.7%(379/418) gay subjects had sex with other men. The median number of the sexual partners was 3, QR=4, and 1 at least and 38 at most. In the control group, 53.0%(175/330) had sex behaviors over the last 6 months. As for the number of sexual partners, the median was 1, QR=1, and the smallest was 1 and the maximum was 22. There was a significant difference between the two groups(Z=-10.878, P=0.000).Up to now,82.3%(344/418) ever had regular sex partners. The median was 3, QR=2. And,53.1%(222/418) had regular sex partners in the last 6 months, and the median is 1, QR=0.(5) Casual sex partners. In the last 6 months,73.0%(305/418) of gay men at least had one casual sex partner; the median was 4, QR=4; 6.7% of them had equal to or more than 10 casual sex partners; and 36.4% of them not only had regular sex partners but also had casual sex partners in the last 6 months. (6) Commercial sex partners:In the latest 6 months,3.8%(16/418) had ever purchased commercial sexual behaviors; 5.3%(22/418) had ever provided commercial sexual behaviors; 6 respondents had unprotected sexual behavior when was purchased commercial sexual behaviors, and 7 respondents had unprotected sexual behaviors when provided commercial sexual behaviors.6. STDs and AIDSIn the past year,73 (17.5%) respondents were infected with at least one kind of STDs, and there were 20 people in control group; there was a signigicant difference (χ2=22.026, P=0.000). Among the 418 gay men, 49.3%(206/418) had ever taken a HIV antibody test; in the past year,159 (38.0%) persons had taken a HIV antibody test, and 4 respondents were proved positive.7. Negative life eventsThe negative life events stimulating quantity of gay men group were scored from 0 to 99. The median score is 10, QR=13(P25=5, P75=18). In the control group, the scores was 0 to 64, and the median was 5, QR=7. Mann-Whitney U test showed that there was a significant statistical difference between the two groups (Z=-5.272, P=0.017), and the gay men's quantity of stimulus from negative life events were higher.Sorted these gay negative life events by danger level, the top five were: "You admitted to yourself that you could be/were gay", "you had moderate financial difficulties", "Your relationship with your lover ended(not because of outside factors like having to leave the country or death)", "You had a minor illness or injury like one needing a visit to the doctor or a couple of days off work", "There have been increasing serious arguments with you your lover"8. Social supportThe scores of SSRS of 418 gay men were ranged from 16 to 54, and the mean was 32.17±6.69. T-test showed that there were significant differences between gay men and straight men (t=-8.914-14.826, P=0.000) in subjective support, objective support and utilization of social supports.9. SuicideTirty point six percent of the gay subjects ever had suicide ideation during the past year, and there were 7.9% in the control group. There was stastistical significant difference between two groups (χ2=58.385, P=0.000). Twenty four (5.7%) gay men had a suicide plan,9 gay men were ready to suicide,10 gay men had ever commit suicide during the past year, and 19 gay respondents (4.5%) were treated because of suicide.Single factor analysis showed that quantity of stimulus from negative life events and coming out are main risk factors of gay men suicide. After controlling other risk factors, these two are still important risk factors, OR was 2.17(P=0.001),1.63(P=0.037)respectively.10. Dating violenceGay men who ever suffered from dating violence was 34.7%, mental violence was 32.3%, physical violence was 7.9%, and sexual violencewas 9.6%. Compared to the control group, there were signigicant differences (Z=-7.543, P=0.000).11. PCS, MCS and related factorsFor SF-36 v2, an extraction procedure was used for main components with VARIMAX command. The scale factor structure illustrated the presence of 2 factors explaining 57.12%% of the total variance; one was PCS, and the other was MCS. Mann-Whitney U test showed that the scores of gay men's PSC and MCS were less than the control group(Z=-7.140,-13.075, P=0.000). After controlling other risk factors(age, education background, original household registration, marital status, history of STDs, suicide ideation, dating violence), there was still a notable difference between the two groups in PCS(OR=8.454, 95%CI:3.983-17.945) and MCS(OR=26.35,95%CI:11.359-61.103).Multifactor retrospective Ordinal analysis showed that sex role in same-sex sexual relations(X3), coming-out(X6), history of STDs(X7), suicide ideation(X9) and dating violence(X10) were the major related factors of PCS. The regression model was as follows:Logit1=-2.583+0.444X3+0.779X6+0.835X7+1.032X9+0.723X1oIn Addition, sex role in same-sex sexual relations(X3),social support(X8),suicide ideation(X9)and dating violence(X10)were the major related factors of MPCS. The regression model was as follows:Logit1=-4.136+0.621X3+0.897X8+1.324X9+1.033X10In total,2202 effective subjects were recruited into this study,1208 men (54.9%) and 994 women (45.1%) whose ages ranged from 16 to 71 years old (M=33.13, SD=16.47); 820(37.2%) came from cities and 1275(57.9%) were rural population, other 107 (4.9%) did not report this data; 182 (8.3%) were less than college,1050 (48.0%) were junior college degree,904 (41.1%) were bachelor degree, and 59 (2.7%) were master or doctor degree; 1806 (82.0%) of the participants were single, 396 (18.0%) were married; 569 (25.8%) were reported to have a job,27 (1.2%) were unemployed, and 1606 (72.9%) were students.The scores of ATLG of all the 2202 straight subjects were ranged from 24 to 99, the mean score was 60.26±12.71. The full range is 75; after dividing the full range into three sections, the results showed that 17.6%respondents'scores were 24-48,69.1%were 49-74, and 13.2% were 75-99.All 2202 subjects'scores of ATG (33.43±7.25) were higher than ATL (26.84±6.64) (t=-40.555, P=0.000);②The male's ATG scores (63.60±12.94) were higher than female's (56.15±11.14) (t=12.638, P＜0.001);③The unmarried subjects'ATLG scores (59.75±12.34) were less than the married subjects'(75.98±14.05) (t=-9.385, P＜0.001).④The rural subjects'ATLG scores (62.64±12.26) were higher than urban subjects'(55.57±13.91) (t=-11.917, P＜0.001).⑤The employed subjects' ATLG scores (68.91±16.03) were higher than unemployed subjects' (59.30±11.91)(t=7.52, P＜0.001);⑥The ATLG scores of those subjects less than college was 65.17±13.30, subjects of junior college degree was 62.38±11.44, and those subjects higher than bachelor degree was (57.89±13.53) (F=28.32, P<0.001);⑦The best ranking item was "I would not be too upset if I learned that my son were a homosexual." (3.85±1.102).Multifactor retrospective Ordinal analysis showed that age (X1), gender (X2), original household registration (X3), marital status (X4), employed or unemployed (X5) and education background (X6) were the major related factors to the attitudes toward lesbians and gay men. The regression model was as follows:Logit1=-7.074+0.431X1+1.195X2+0.297X3+1.680X4+0.833X5+0.818 X6In total,701 effective subjects were recruited into this study,; they were consisted of two parts:①404 college students:210 men and 194 women whose ages ranged from 17 to 25 years old(M=19.66, SD= 1.03); 10 days after the first test,193 students were re-tested;②297 community residents:144 men and 153 women whose ages ranged from 18 to 71 years old(M=45.54, SD=15.24).The test-retest coefficients of subscales were 0.941,0.952, and of total-scale,0.967; Correlation coefficients between each item and the total ATLG score ranged from 0.415 to 0.738; Cronbach's a coefficient of the total-scale and two sub-scales were 0.913,0.867 and 0.879, respectively; the split-half reliability of the scales ranged from 0.868 to 0.934.The ATLG scale factor structure illustrated the presence of 3 factors explaining 53.788% of the total variance. The items 1,3,5,6,8,9,10 were the first cluster of factors, which conceptually represented beliefs toward lesbians. All items of the ATG scale were the second cluster of factors, which conceptually represented beliefs toward gay men. The items 2,4,7 were the third cluster of factors, which conceptually represented social sanction. For the ATL subscale, the factor structure showed one main factor that would explain 46.221% of the variance, which conceptually represented beliefs toward lesbians. And for the ATG subscale, the factor structure also showed one main factor that would explain 48.676% of the variance, which conceptually represented beliefs toward gay men.Conclusions:1. Gay men's dating relationships are unstable and short duration, and most gay men have no plan for their future when they fall in love; most gay men will marry a heterosexual woman, and want to have children; the form marriage (a gay marry a lesbian) has became a new marriage way in gay men population.2. Gay men are even more likely to look for sex partners through internet; the sexual relationships are unstable; dangerous sex behavior are common; and there are a great possibility to be infected with STDs and AIDS.3. The negative life events stimulating quantity of gay men are great; there are higher rates of dating violence; they have low social support; and suicide has become a serious social problem among them.4. The Gay men's living conditions are relatively poor, and the level of their physiological and psychological health conditions are relatively low. Playing a passive role in homosexual relations, suffering from STDs, suicide ideation and dating violence were main risk factors to gay men's poor PCS; playing an active role in homosexual relations, higher SSRS, no suicide ideation and dating violence were main protective factors to gay men's poor MCS.5. Seventeen point six percent of publics hold positive attitudes toward lesbians and gay men; all publics hold more negative attitudes toward gay men than lesbians; for all subjects, the most intolerable thing is their own son is a gay.6. The attitudes toward lesbians and gay men were related to age, gender, original household registration, marital status, employment status, and education background.7. The Attitudes toward Lesbians and Gay Men Scale designed by Gregory M. Herek has a good reliability and validity in Chinese, and can be used in research of attitudes toward lesbians and gay men in the social cultural context of China.Significances and innovationsThis research has taken gay men in Changsha as the research subjects. A cross-sectional and case-control study was conducted to describ their sociodemographic features, romantic relationships, features of their sexual behaviors, suicide, life events, social support, dating-violence and quality of life clearly and detailedly. The scientific basis has been provided to improve the gay men's living conditions and take a more targeted crisis intervention。The study recruited samples through internet randomly, and 418 effective samples were included at last. Research about gay men are very difficult, and this study has provided a new way to solve this problem. It's a totally new try of methodology.This study was the first attempt to test the reliability and validity of Herek's Attitudes toward Lesbians and Gay Men Scale (ATLG) in China. It provided a more scientific tool to study the attitudes toward homosexuality. Then, we used the ATLG scale to assess the public's attitudes toward lesbians and gay men.