Background

Transgender is a definition of how a person feels about their gender identity [1]. A person is cisgender when their gender at birth matches their assigned gender and transgender when their gender at birth does not match their assigned gender [1]. Transgender women are individuals who were assigned male at birth but identify as female [2, 3]. In addition to psychological gender identity, transgender people may undergo sex-change surgery, take hormones (contraceptives, etc.), and inject their bodies with fillers to change their characteristics to make their bodies conform to their gender identity. Because of the diversity of the community of transgender women, they may not change their appearance and live instead in male-presenting bodies [1].

Transgender women are one of the groups with the highest infection rate of human immunodeficiency virus (HIV) in the world, which has characteristics of concealment and a low detection rate. In a global study of the burden of AIDS (acquired immune deficiency syndrome), the odds of HIV infection among transgender women was 48.8 times that of cisgender adults of reproductive age [4]. Additionally, a meta-analysis showed that the HIV-positive rate among transgender women in 15 countries was 19.1%, the HIV prevalence rate among transgender women in low- and middle-income countries was 17.7%, and the HIV prevalence rate in high-income countries was 21.6% [4]. Transgender women exhibited more high-risk behaviors, such as drug abuse, multiple sexual partners, unprotected sex, etc. [5]. The prevalence rate of HIV among transgender women was significantly higher than that among men who have sex with men (MSM). According to the gender minority stress model of Mayer and others [6], transgender women face greater stress, are more vulnerable to high levels of discrimination and violence, and face a higher risk of HIV infection and transmission [7].In addition, because the gender division is still dualistic, transgender people are excluded from most social infrastructures [8]. A study in the United States reported that transgender people who self-reported falling outside the binary category had worse health outcomes, including mental health, than their binary peers [9]. This increases the risks faced by minority groups, such as transgender people, who already face disadvantages such as discrimination and violence [10]. Research on transgender people showed that they were more likely than cisgender people to be diagnosed with depression and substance use disorders, engage in self-harm and die by suicide [10, 11]. Transgender individuals have worse mental health than cisgender individuals [9, 12]. However, there is little research on transgender women in China, especially on the factors that influence psychology. A study has pointed out that when seeking social support, individuals will choose to make changes to more positive behaviors and find an alternative way to obtain happiness and make a promotional response [13]. There was a moderate correlation between social support and physical and mental health [13, 21], while in this study, we used face-to-face interviews and filled out paper questionnaires to ensure quality. After the questionnaire was completed, participants received a free HIV/AIDS syphilis test provided by us. Among the 500 questionnaires collected, 498 were completed effectively after checking the completeness and consistency of the questionnaires and the qualified questionnaires were coded, and the questionnaire efficiency rate was 99.6%. To ensure the accuracy of the data, double entry was performed.

Measures

12-item General Health Questionnaire (GHQ-12)

For psychological distress, the 12-item General Health Questionnaire (GHQ-12) was used. The GHQ is one of the most widely used questionnaires for screening mental disorders and has been verified by many studies and has good reliability and validity. The English version has been translated into many languages and used in countries with different cultural backgrounds [22, 23]. There are 60-, 30-, 28-, 20- and 12-items of versions of the general health questionnaire [24]. The 12-item General Health Questionnaire (12-item General Health Questionnaire, GHQ-12) was used in this study. It has been verified many times and is one of the most commonly used measurement tools for mental health problems. It has the advantages of simplicity, time savings and fast response.

According to the research of Yang [24], it has been verified that the Chinese version of Taiwan province and the GHQ-12 have good applicability in mainland people, and the screening efficiency of the GHQ is similar to that of the Symptom Check List-90 (SCL-90). Cronbach's α was 0.92 in this study. The scoring method used in this study was the GHQ standard scoring method [29]. Additionally, the Pearson chi-square value in the chi-square test was consistent with the chi-square score in binary logistic regression analysis.

Before multivariate analysis, we tested the collinearity among variables and found that the variance inflation factor (VIF) of each variable was less than 10. In univariate logistic regression results, variables with p < 0.20 were incorporated into the multivariate analysis model. All comparisons were bilateral, and p < 0.05 was considered to be statistically significant in multivariate logistic regression.

Results

GHQ—12 positive rate

The critical value of GHQ-12 positive rate is 3. If the score is greater than or equal to 3, the psychological state is negative. A negative psychological state is a positive result (psychological distress), that is, there may be a psychological problem. In 498 participants, 1 in 5 transgender women suffered from psychological distress (20.08%), and the average GHQ-12 score was 1.33 (SD = 1.47).

Factors influencing psychology by social demographic characteristics

Among 498 participants, the mean age was 29.58 years (SD=5.73). Table 1 shows the number and proportion of transgender women who had psychological distress in terms of sociodemographic characteristics. Almost one quarter (22.7%) of unmarried transgender women had psychological distress compared with only 14.3% among those who were married, and only 6.7% among those who were divorced or widowed. Regarding the province of origin of the transgender women, 18.7% of those from Shandong had psychological distress compared with 22.2% among transgender women from other provinces. Two fifths (39.8%) of transgender women were from other provinces which indicated that they were highly mobile. During the investigation, according to the interviewees' self-report, in addition to the relatively strong mobility of transgender women engaged in relevant stage performances, some participants engaged in sexual services across provinces or cities according to the needs of customers. In terms of transgender women’s residence and occupation, there was a higher prevalence of distress among those living in towns and cities (around 20%) compared with those living in rural areas (15.3%) and among those working in the service industry (23.5%) compared with those in other occupations (10.9%).

The service industry included waiters, liquor salesmen, performers, cashiers, and sex workers, and other occupations included administrators, teachers, soldiers and students. Considerably lower prevalence among those with tertiary education. The univariate logistic regression results showed that education was a factor related to psychological distress. There was an inverse relationship between monthly income and prevalence of distress.

The result of univariate logistic regression analysis showed that individuals aged between 30 and 39 years (OR:0.37, 95% CI:0.0.22-0.60) and individuals over 39 years of age (OR:0.17, 95% CI:0.05-0.57) were less likely to have psychological distress than transgender women 18-29 years of age. In addition, transgender women who were divorced or widowed (OR: 0.24, 95% CI: 0.09-0.69) were less likely to have psychological distress than those who were unmarried. Transgender women who were engaged in occupations outside the service industry (OR:0.40, 95% CI:0.22-0.72) and those who earned more than 15,000 yuan (OR:0.54, 95% CI:0.04-0.87) were less likely to have psychological problems (Table 1).

Table 1 Univariate logistic regression analysis of sociodemographic characteristics related to transgender women's psychological distress
Full size table

AIDS-related knowledge

The AIDS-related knowledge level was used to represent cognition on AIDS. The correct rate of “HIV-infected women will give birth to children infected with AIDS?” was the lowest (90.16%), and the correct rate of “Is there any legal responsibility for deliberately spreading AIDS?” and “Will transfusing blood with HIV infect or spread AIDS?” was highest (99%). The correct rate of each topic is shown in Supplementary Fig. 1.

Univariate logistic regression analysis indicated that the AIDS-related cognitive status was not significant (p > 0.2) and AIDS-related cognitive status was not a related factor of psychological stress. (Table 2).

Table 2 Univariate logistic regression of AIDS-related cognition, sexual behavior in six months, substance abuse, social support and STIs
Full size table

Related sexual behavior, substance abuse and social support

From the number of anal intercourse in the previous week, only 12.1% of transgender women who had anal sex more than ten times last week had psychological distress compared with around 20% among those having anal sex less frequently. Among the transgender women who wore condoms during each anal sex encounter, 24.2% had psychological distress, compared with only 14.7% among those who did not consistently wear condoms during anal intercourse. In terms of frequency of oral sex condom use, 37.2% of transgender women who wore condoms every time reported psychological distress, while only 18.1% of the others did so. Transgender women who had fewer than three sexual partners within six months had no psychological distress, while 17.1% of transgender women with three to five sexual partners had psychological distress, and 21.1% of transgender women with more than five sexual partners within six months had psychological distress.The prevalence of psychological stress was similar among transgender women who drank and did not drink alcohol but was higher among those who used drugs (25.7%) than among those who did not (16.7%). In addition, we also included a history of sexually transmitted infections (STIs) other than AIDS in the univariate logistic regression analysis. Only 12.1% of transgender women with STIs had psychological distress, while 24.7% of transgender women without STIs had psychological distress. The average score for social support was 33 (SD = 9.67). Among transgender women with insufficient and general social support, 22.5% and 27.3% had psychological distress, while among transgender women reporting high social support, only 15.7% had psychological distress. We found that using condoms every time during anal sex (OR:1.85, 95% CI:1.16–2.94), using condoms every time during oral sex (OR:2.68, 95% CI:1.45–4.97), drug use (OR:1.72, 95% CI:1.10–2.68), STIs (OR:0.42, 95% CI:0.25–0.70), and scores between 20 and 30 on social support (OR:2.01, 95% CI:1.27–3.21) were all factors related to the psychological distress of transgender women. (Table 2).

The relation of gender identity and adolescent experience with psychological distress

We also performed some analysis on gender identity and the psychological impact of adolescent experiences on transgender women. These aspects were reflected in some problems, and were also included in the univariate logistic regression analysis, with the specific topic as follows: “Did you ever think that being a girl was better than being a boy when you were a teenager?”, “Did you particularly like to play with boys when you were child?”, “Do you agree with your own sexual orientation?”, “Have you made your sexual orientation and identity public?”, “Do you want to be recognized for your sexual orientation and identity?”, “Do you think the people around you think differently about you?” and “Has the homosexual culture on the internet helped you determine your sexual orientation?”.

The univariate logistic regression analysis showed that the factors that related to psychological distress. Compared to those who thought that being a girl was better than being a boy at an early age, those who did not have such thoughts were less likely to have psychological distress (OR: 0.33, 95% CI: 0.18–0.62). Other characteristics of transgender women with lower prevalence of psychological distress were: never disclosing their sexual orientation and identity, or only disclosing to friends and family (OR: 0.22, 95% CI: 0.14–0.40, and OR: 0.22, 95% CI: 0.10–0.48, respectively); and not feeling having strange eyes around (OR: 0.28, 95% CI: 0.15–0.51). On the contrary, there were other factors that might increase the risk of psychological distress of transgender women, which included not being raised as a girl at an early age (OR:1.41, 95%CI:0.90–2.23), not desiring to be identified with one's sexual orientation and identity (OR:9.61, 95% CI:5.03–18.35), and the Internet not playing an essential role in hel** determine sexual orientation (OR:12.17, 95% CI:7.09–20.93). (Table 3).

Table 3 Results of univariate logistic regression of gender identity and adolescent experience on psychological distress
Full size table

Predictors of transgender women's psychology

Significant variables in univariate logistic regression were incorporated into the multivariate logistic regression model (p < 0.20). Multivariate logistic regression analysis showed participants who earned a monthly income between 10,000–15,000 yuan (OR: 0.16, 95% CI:0.06–0.45) or greater than 15,000 yuan (OR: 0.07, 95% CI:0.01–0.43) were less likely to have psychological distress. Additionally, transgender women who never disclosed their sexual orientation and identity (OR: 0.19, 95% CI: 0.06–0.58), who only disclosed their sexual orientation and identity to families and friends (OR: 0.41, 95% CI: 0.18–0.93) were also less likely to have psychological distress. By contrast, transgender women who did not desire to be identified with their sexual orientation and identity (OR: 3.31, 95% CI: 1.08–10.16) and who reported that the Internet did not play an essential role in hel** determine sexual orientation (OR: 5.96, 95% CI: 2.91–12.20) were more likely to have psychological distress. (Table 4).

Table 4 Predictors of transgender women's psychological distress in multivariate logistic regression
Full size table

Discussion

To the best of our knowledge, this study is the first to use the GHQ-12 in Shandong province, China, to investigate the psychological factors affecting transgender women. In this study, we added a new variable of high-risk sexual behavior, that is, the frequency of oral condom use. Among other psychologically related factors, we also included the influence of the internet on transgender women. The majority of participants were young people under 39 years old (90.6%). A previous study showed that the psychology of transgender women was related to their age, and participants of different ages showed obvious mental health differences [30]. In our univariate logistic regression study, younger age was one of the factors related to the psychological distress of transgender women. Among the transgender women, a majority were working in the service industry (72.5%). Some of the transgender women who were unemployed would selectively serve or be kept by a client. In addition, the mobility of transgender women in this study was relatively strong, with 39.8% of nonlocal transgender women [31]. Although this study did not find a relationship between the mobility of transgender women and their psychological distress, factors such as the high mobility of transgender women, the burden of the HIV epidemic and access to resources deserve further study [32].

Related studies have shown that the differences between urban and rural areas are related to the psychological state of LGBT individuals [33, 34]. There are many reasons for the association, which may be because transgender women living in cities and towns can access public services more easily and are more vulnerable to discrimination and pressure from the outside [35, 36]. However, we did not find an association between psychological distress and residence in our analysis (p > 0.2).

Sex services and high-risk sex behaviors were common among transgender women, and sex work has been reported to be associated with more stigma and discrimination [37]. On the basis of the description of the transgender women surveyed, they were engaged in general service industries at the same time, such as performance and beverage sales, and also performed sex services part-time. Some held part-time jobs in the general service industry, while sex work was their main occupation. Although occupation of transgender women was a factor associated with psychological distress in univariate logistic regression, occupation was not significant in multivariate logistic regression. For transgender women who had psychological distress, their perception of unprotected sexual behavior may be affected [38]. At present, research on psychological distress and high-risk sexual behavior generally includes the influence of psychological status on high-risk sexual behavior, or whether sexual abstinence and risky sex are associated with psychological symptoms [39, 40]. In our analysis, we regarded high-risk sexual behavior as a factor related to the psychology of transgender women. When asked whether they wear condoms during anal sex, only 56.4% of transgender women reported wearing condoms every time they had sex. We found that the frequency of wearing condoms during anal sex was an important factor related to the psychological distress of transgender women in the univariate logistic regression analysis. In the survey, we learned that some transgender women engaged in sexual services, and those transgender women gave up wearing condoms because customers provided more money or other benefits. However, in the multivariate model, the frequency of anal condom use was not significant. Similarly, the frequency of oral condom use was significant in univariate logistic regression but when included in multivariate logistic regression it was no longer significant. Even so, since there are few studies on oral sex, especially among transgender women, the oral sex behavior of trans women deserves further research. In our survey, according to the statements of most transgender women, who believed that it was almost impossible to transmit AIDS or other sexually transmitted diseases through oral sex, all generally chose not to wear condoms. They thought that unless their oral cavity was damaged, they could not contract AIDS. On the other hand, they believed that this would affect the comfort of oral sex. A study on oral sex among transgender youth pointed out that although the surveyed youth had some knowledge about oral sex, they chose not to wear a condom because they did not want to wear them, they thought condoms felt uncomfortable or they did not think they needed to wear them [41]. The frequency of wearing condoms during oral sex was very low, and only 10.2% of transgender women chose to wear condoms every time they engaged in oral sex. Of the transgender women who wore condoms every time they had oral sex, 37.2% reported psychological distress. Therefore, it is still an important aspect to formulate individualized intervention measures, especially to improve the communication between transgender women and partners who choose to wear condoms during sexual intercourse.

The monthly income of transgender women was a factor that affected the psychological distress of transgender women. Compared to transgender women with a monthly income of less than 5000 yuan, transgender women with an income of 10,000–15,000 (OR: 0.16, 95% CI: 0.06–0.45) and with an income of more than 15,000 (OR: 0.07, 95% CI: 0.01–0.43) were less likely to have psychological distress. A study from the US found that transgender people with low incomes report more psychological distress [42]. A study in Taiwan province showed that LGBT groups with incomes above NT$ 5,000 (equivalent to 4,500 yuan) had better physical health, psychological health, social relationships and environmental health [43]. This portion of high-income transgender women might be due to the improvement in the quality of life brought about by the improvement in economic conditions while increasing the sense of security in personal life.

Transgender women in China are a vulnerable group. Due to their traditional cultural background, they receive less social support. Many transgender women found that it was difficult to choose to disclose their identity and obtain social support. Legitimate gender identity may improve the psychological status of transgender women [44, 45]. Inconsistency in identity and behavior can lead to adverse health consequences, including psychological problems such as anxiety, depression, suicide, drug abuse, and high-risk sexual behaviors [42, 46]. Sexual orientation is also risk factor for the spread of AIDS [47]. Compared with disclosing their sexual orientation and identity, transgender women who did not disclose their sexual orientation (OR: 0.19, 95% CI: 0.06–0.58) and identity or only disclosed their sexual orientation and identity to families and friends (OR: 0.41, 95% CI: 0.18–0.93) were less likely to have psychological distress. Due to the existence of the traditional concept of dual gender, they were afraid of discrimination brought by different views, causing their anxiety symptoms [48]. The research on the gender identity of transgender women has always been a problem worth paying attention to. In a qualitative study of transgender women in Jiangsu province, China, most transgender women only tended to reveal their gender identity to family members or close friends, but generally, in the family, they will be rejected by their family and may be subjected to violence or abuse [

Conclusion

1 in 5 transgender women suffered from psychological distress (20.08%). As a vulnerable group, transgender women are prone to psychological distress. In summary, the GHQ-12 was used to measure the psychological factors affecting transgender women in Shandong, China. Having a higher income were better for transgender women’s psychological health.

Although the included variables of high-risk sexual behavior were not predictors of the psychological distress of transgender women, an association was found in univariate logistic regression analysis. Transgender women who chose to come out were more likely to report psychological distress than those who refused to disclose their sexual orientation and gender. Improving the internet's understanding of transgender women's gender, sexual orientation and role will help to spread common knowledge about transgender women and further improve the social acceptance of transgender women. Therefore, we suggest formulating appropriate strategies combined with the internet to influence the sexual behavior of transgender women, improve their social understanding and pay attention to their psychological health.