App use frequency and condomless anal intercourse among men who have sex with men in Beijing,China: a cross-sectional study

Abstract

Gay dating applications (apps) are popular among men who have sex with men (MSM) in China. This study examined the relationship between gay dating app use frequency and condomless anal intercourse (CAI) in a sample of MSM in Beijing. In October 2017, a web-based, cross-sectional survey was conducted among MSM from Beijing, China. Demographics, frequency of app use, and sexual behaviors were collected and analyzed. Multivariable logistic regression and restricted cubic splines were used to evaluate the association between app use frequency and CAI. Of the 9280 participants, the proportion of CAI within one month prior to the survey was 43.3%, and 43.7% logged onto the app ≥11 times per week. In adjusted logistic regression analysis, logging onto the app 6–10 times per week and ≥11 times per week were inversely associated with CAI (adjusted odds ratio [AOR] =0.88 [95%CI, 0.78–0.99]; AOR = 0.88 [95%CI, 0.82–0.97], respectively). Among those 40 years of age or older, logging onto the app ≥11 times per week was associated with lower rates of CAI (AOR = 0.57 [95%CI, 0.40–0.81]). Results indicate that frequent app use was associated with lower odds of CAI among MSM in Beijing, China. Innovative interventions, which take advantage of this popular media platform, should be developed as such apps have been found to reduce the odds of CAI in Beijing.

Keywords

HIV sexual behavior homosexual high-risk behavior sexual intercourse

Introduction

The human immunodeficiency virus (HIV) infection epidemic causing acquired immune deficiency syndrome (AIDS) among men who have sex with men (MSM) continues to increase across the world.¹ Systematic reviews and meta-analyses have reported an increasing prevalence among MSM in China from 1.4% in 2001 to 5.3% in 2009.² This prevalence has continued to rise over the past decade, from 5.73% in 2010 to 8% in 2015.^3,4 According to the Chinese Center for Disease Control and Prevention (China CDC), more than 25% of all newly diagnosed HIV infections were among MSM in 2017.⁵ In larger, urban areas of China, MSM have been found to account for more than 50% of all newly diagnosed HIV cases. MSM accounted for 54% of new diagnoses in southeastern Zhejiang province, 74% in Hebei province, and 76% in Tianjin city in northern China.⁶ Moreover, data from the National HIV/AIDS Case Reporting System showed that MSM in the capital city of Beijing were responsible for a larger portion of new cases, accounting for almost 78% of all newly diagnosed HIV infections in 2017, ranking number 1 both in terms of proportion and absolute number of HIV infection by MSM among 31 provinces, municipalities, and autonomous regions.^7,8

MSM are at higher risk for HIV transmission because they are heavily stigmatized due to their homosexuality,⁹ which makes them less likely to disclose their sexual orientation, particularly to healthcare professionals.¹⁰ In response, nongovernmental organizations (NGOs), which collaborate with local CDC offices in different cities throughout China, provide the much necessary HIV prevention work to the MSM community.¹¹ While these NGOs have played an important role in providing necessary outreach and care, they often lack the structural, financial, and professional support required to meet the needs of MSM throughout China. Subsequently, it is more difficult for conventional, face-to-face interventions to reach this at-risk population. This in turn reduces MSM’s access to HIV-related health services.¹² As such, understanding the epidemic drivers, including high-risk sexual behaviors, of HIV transmission is necessary in order to develop innovative strategies and improve health services to reduce the spread of HIV among this group.

Over the past years, how MSM seek sexual partners has drastically changed. MSM have moved away from the traditional ways of meeting partners, such as public spaces like bars, to seeking sexual partners through geosocial networking (GSN) applications, particularly gay dating applications (apps). For example, one of the most popular gay-dating apps, Grindr, had more than four million users across 196 countries in 2018, and more than 3 million users logged onto the app every day.¹³ Blued (Blued City Holdings, Ltd), the most popular gay dating app for Chinese MSM specifically, had 480,000 active users monthly in Beijing alone, in 2018.¹⁴ This increasing popularity of gay dating apps has created new opportunities for finding sexual partners¹⁵ and making friends within the gay community.¹⁶ One recent study found that in China, approximately 40% of MSM have sought male sexual partners by using such apps.¹⁷

Given the high HIV burden and the popularity of gay dating app usage among MSM, it is necessary to understand mobile app use and its potential influence on MSM sexual behaviors. Some studies have reported that gay dating app usage was associated with an increase in condomless anal intercourse (CAI),^18,19 while other studies indicated there were no statistical relationships between app use and CAI.^17,20 However, most existing studies have mainly analyzed high-risk sexual behaviors between gay dating app and non-app users²¹ and do not consider the frequency of app use. Moreover, the majority of these studies were conducted in high-income countries,¹⁸ with different socio-cultural perspectives of relationships and partnership seeking compared to China. In China, due to cultural expectations, MSM often marry into heterosexual partnerships in order to hide their homosexuality.²² Studies have estimated that between 50 and 70% of MSM in China have had sex with females and that up to 90% will eventually marry a partner of the opposite sex.²¹ As such, the impact of GSN app use to discreetly find partners makes this a unique population to study. We aimed to address this gap in the literature by exploring the relationship between gay dating app use frequency and CAI among MSM app users in Beijing, China, home to the largest number of MSM Blued app users.²³

Methods

Recruitment

We conducted a cross-sectional study in October 2017 in Beijing, China. Participants were recruited through Blued, the most popular gay dating app for Chinese MSM.^14,23 Personal invitation messages including a brief introduction of the study and a survey link were privately sent to potential participants from a Blued administrator account. Participants could self-select into the study by clicking on the link. Eligibility criteria for this cross-sectional study included: (1) at least 18 years of age, (2) resided in Beijing, defined as both registration Internet Protocol (IP) and most recent login IP before the survey, (3) logged onto the app at least once in the past week before enrollment, (4) self-reported having had anal intercourse with another man in the six months preceding enrollment, and (5) self-reported negative or unknown HIV status. All participants read informed consent agreements and provided signed electronic informed consent before starting the survey. All participants could email or call the study implementers regarding questions about the study if necessary. Upon completion, participants received a cash coupon worth 40 RMB (about US$ 6) for their time participating in the study. This study was approved by the International Review Board of the National Center for HIV/STD Control and Prevention, Chinese Center for Disease and Prevention (approval number: X170623469).

Data collection

The method mentioned by Broder et al.²⁴ for questionnaire development was used to create the initial survey. Fifteen MSM who worked at Blued reviewed the survey and provided feedback. An additional 50 Blued app-using MSM were conveniently selected to respond to the survey after modification. The final survey questionnaire was generated in July 2017 (available upon request). Participants responded to the survey in October 2017.

Demographics

Socio-demographic variables in the survey were self-reported and included age, race/ethnicity, education level, marriage, and hukou. Hukou refers to China’s national household registration system, determined by where the participant was born.²⁵ Sexual behaviors collected in the survey included sexual debut with a man, sexual orientation, and perception of HIV risk. Sexual debut with a man was divided into three categories: <19 years old, 19–29 years old, ≥30 years old. Sexual orientation was defined as ‘gay,’ ‘straight,’ or ‘bisexual.’²⁶ Perception of HIV risk in Beijing was a binary variable defined as ‘not serious’ or ‘serious.’ HIV testing history was also collected and defined according to the HIV testing guidelines for MSM: ‘never conducted an HIV test,’ ‘conducted an HIV test 1 year before,’ or ‘conducted an HIV test within 1 year.’

Frequency of app use

Two variables were collected: frequency of logging onto the app per week and time since registering on the app. The frequency of logging onto the app per week was assessed by asking ‘how often do you log onto the Blued app per week in your daily life?’ Because different categories of app login frequency have been reported^18,27 in our study, three categorical options were defined based on our pre-survey expert consultation: ‘≤5 times per week,’ ‘6 to10 times per week,’ and ‘≥11 times per week.’ The variable ‘how long have you been registered on the app?’ was a binary variable with two categorical options ‘<1 year’ and ‘≥1 year.’

CAI

The outcome of interest in the study was CAI, assessed by asking ‘In the past month, how often did you use condoms for anal sex with a man?’ with four response options, ‘never use condoms,’ ‘sometimes use condoms,’ ‘often use condoms,’ and ‘always use condoms.’ Since MSM who never use condoms have a similar risk of HIV infection compared with those who sometimes or often use condoms,²⁸ the first three options were coded as ‘CAI,’ and the last as ‘non-CAI.’

Statistical analysis

Statistical analyses were performed by SAS software, version 9.3 (SAS Institute, Inc., USA). Descriptive analysis was used to report the percent distribution of categorical variables and Chi square tests or Fisher’s exact tests were used to analyze differences in demographic, sexual behaviors, and app use across different CAI groups. To explore the relationship between the frequency of app use and CAI, a multivariable logistic regression analysis was used to calculate adjusted odds ratios (AORs) and 95% confidence intervals (CIs), with p < 0.05 considered as statistically significant. Statistically significant variables in the Chi square test and potentially important variables according to directed acyclic graph²⁹ were included in the multivariate logistic regression model. The interaction of age and app use frequency on CAI was also considered. Further, restricted cubic splines (RCSs)³⁰ incorporated in the multivariate logistic regression model were used to assess the dose–response association between the frequency of app use and CAI.

Results

Study participants

The participant enrollment process is shown in Figure 1. A total of 29,307 men opened the survey link, 9280 completed the survey and were included in the study, for an overall completion rate of 31.66%.

Figure 1.

The participant recruitment process.

A summary of demographic and behavioral characteristics can be found in Table 1. A majority of participants were aged <40 years (91.52%), of Han Chinese ethnicity (92.22%), completed college or higher education (68.08%), were single (79.34%), and non-Beijing hukou (63.21%).

Table 1.

Socio-demographic and behavioral characteristics of Blued app-using men who have sex with men in Beijing, China in 2017 (N = 9280).

Variables	N (%)
Age (years)
<40	8493 (91.52)
≥40	787 (8.48)
Ethnicity/race
Han	8558 (92.22)
Others	722 (7.78)
Education
College or above	6318 (68.08)
Senior high	2071 (22.32)
Junior high or below	891 (9.60)
Marriage
Single	7363 (79.34)
Married	1593 (17.17)
Divorced/Widow	324 (3.49)
Registered residence/Hukou
Other	5866 (63.21)
Beijing	3414 (36.79)
Login app frequency (times/week)
≤5	3523 (38.06)
6–10	1697 (18.29)
≥11	4051 (43.65)
Time since registering on the app
≥1 year	7578 (81.66)
<1 year	1702 (18.34)
Perception of HIV risk in Beijing
Serious	1947 (20.98)
Not serious	7333 (79.02)
Sexual debut with a man (age in years)
19–29	5899 (63.57)
<19	2594 (27.95)
≥30	787 (8.48)
Most recent HIV test
≤1 year	5012 (54.01)
>1 year	1007 (10.85)
Never	3261 (35.14)
CAI
Yes	4018 (43.30)
No	5262 (56.70)

CAI: condomless anal intercourse; HIV: human immunodeficiency virus.

Among respondents, a little less than half logged onto the app ≥11 times per week (43.65%) and more than 80% (81.66%) registered on the app more than one year ago. Additionally, a series of sexual and HIV risk behaviors were assessed. More than 60% experienced their first homosexual activity between 19 and 29 years old (63.57%) and regarded the HIV epidemic in Beijing as not serious (79.02%). More than 30% of participants had never received an HIV test (35.14%) and 43.3% experienced CAI in the past month before survey.

Correlates of CAI

Men with CAI were more likely to be aged ≥40 years old (10.25% versus 7.13%, χ² = 28.71, P < 0.001) and to be less educated (P < 0.001). Men with CAI were less likely to be single (76.48% versus 81.51%, χ² = 37.01, P < 0.001); from Beijing hukou (35.42% versus 37.87%, χ² = 5.75, P = 0.02); and were more frequent app users, logging onto the app ≥11 times per week (41.76% versus 45.09%, χ² = 11.15, P = 0.01). Additionally, men with CAI were less likely to have registered on the app ≥1 year ago (78.75% versus 83.89%, χ² = 40.17, P < 0.001). Men reporting CAI, however, were more likely to have experienced their first homosexual activity between 19 and 29 years old (58.61% versus 67.33%, χ² = 77.07, P < 0.001), perceived the HIV epidemic in Beijing as not serious (19.59% versus 22.06%, χ² = 8.3, P = 0.004), and had an HIV test within one year prior to the survey (49.58% versus 57.44%, χ² = 68.24, P < 0.001) (Table 2).

Table 2.

Comparison of demographics, app use, and behavioral characteristics between Blued app-using men who have sex with men reporting condomless anal intercourse and those reporting condom-protected anal intercourse in Beijing, China in 2017 (N = 9280).

Variables	CAI (n, %)	Non-CAI^a (n, %)	Chi square	P value^b
Total	4018 (100)	5258 (100)
Age (years)			28.71	<0.001
<40	3606 (89.75)	4883 (92.87)
≥40	412 (10.25)	375 (7.13)
Ethnicity			0.79	0.37
Han	3694 (91.94)	4860 (92.43)
Others	324 (8.06)	398 (7.57)
Education			403.39	<0.001
Junior high or below	545 (13.56)	346 (6.58)
Senior high	1182 (29.42)	889 (16.91)
College or above	2291 (57.02)	4023 (76.51)
Marriage			37.01	<0.001
Single	3073 (76.48)	4286 (81.51)
Married	775 (19.29)	818 (15.56)
Divorced/Widow	170 (4.23)	154 (2.93)
Registered residence/Hukou			5.75	0.02
Beijing	1423 (35.42)	1991 (37.87)
Others	2595 (64.58)	3267 (62.13)
App login frequency (times/week)			11.15	0.01
≤5	1590 (39.57)	1940 (36.90)
6–10	750 (18.67)	947 (18.01)
≥11	1678 (41.76)	2371 (45.09)
Time since registering on the app (years)			40.17	<0.001
<1	854 (21.25)	847 (16.11)
≥1	3164 (78.75)	4411 (83.89)
Sexual debut with a man (age in years)			77.07	<0.001
<19	1259 (31.33)	1335 (25.39)
19–29	2355 (58.61)	3540 (67.33)
≥30	404 (10.05)	383 (7.28)
Perception of HIV risk in Beijing			8.30	0.004
Serious	3231 (80.41)	4098 (77.94)
Not serious	787 (19.59)	1160 (22.06)
Most recent HIV test			68.24	<0.001
Never	1596 (39.72 )	1661 (31.59)
>1 year	430 (10.70)	577 (10.97)
Within one year	1992 (49.58)	3020 (57.44)

CAI: condomless anal intercourse; HIV: human immunodeficiency virus.

Four participants were excluded because they reported not having anal intercourse in the past month.

p < .05 denotes statistically significant difference within the row.

Multivariable analyses of frequency of logging onto the app and CAI

In the adjusted logistic regression model, CAI was significantly associated with the frequency of app use. MSM who logged onto the app 6–10 or ≥11 times per week were 12% less likely to have CAI compared with MSM who logged onto the app ≤5 times per week (AOR = 0.88, 95%CI: 0.78–0.99 for frequency of app use 6–10 times/week; AOR = 0.88, 95%CI: 0.80–0.97 for frequency of app use ≥11 times/week) after adjusting for potential confounders (Table 3).

Table 3.

Associations between the frequency of Blued app-use and condomless anal intercourse among men who have sex with men in Beijing, China in 2017 (N = 9280).

	All study participants (N=9280)		Age <40 years old (n=8493)		Age ≥40 years old (n=787)
Variables	CAI (%)	AOR (95%CI)	CAI (%)	AOR (95%CI)	CAI (%)	AOR (95%CI)
Age (years)
<40	42.46	Ref	–	–	–	–
≥40	52.35	1.28^a (1.08, 1.52)	–	–	–	–
Education
College or above	36.26	Ref	35.78	Ref	43.06	Ref
Senior high	57.07	2.17^a (1.96, 2.41)	56.49	2.20^c (1.97, 2.45)	61.99	2.09^c (1.47, 2.98)
Junior high or below	61.17	2.45^a (2.11, 2.86)	60.57	2.51^c (2.13, 2.96)	64.19	2.17^c (1.42, 3.31)
Marriage
Single	41.74	Ref	41.63	Ref	45.55	Ref
Married	48.65	1.03 (0.91, 1.17)	46.56	1.03 (0.90, 1.18)	53.39	1.12 (0.78, 1.61)
Divorced/Widow	52.47	1.25 (0.98, 1.58)	48.84	1.12 (0.84, 1.48)	59.63	1.72^b (1.04, 2.84)
Registered residence/Hukou
Other	44.24	Ref	43.43	Ref	54.42	Ref
Beijing	41.68	0.98 (0.90, 1.07)	40.73	0.99 (0.90, 1.08)	49.86	0.91 (0.68, 1.23)
App login frequency (times/week)
≤5	45.02	Ref	44.10	Ref	58.33	Ref
6–10	44.20	0.88^b (0.78, 0.99)	42.61	0.88 (0.78, 1.00)	56.91	0.73 (0.48, 1.10)
≥11	41.42	0.88^a (0.80, 0.97)	41.76	0.92 (0.83, 1.01)	46.36	0.57^c (0.40, 0.81)
Time since registering on the app
≥1 year	41.75	Ref	41.05	Ref	49.52	Ref
<1 year	50.18	1.11 (1.00, 1.25)	48.80	1.09 (0.97, 1.23)	63.52	1.43 (0.98, 2.10)
Perception of HIV risk in Beijing
Not serious	44.06	Ref	43.11	Ref	53.97	Ref
Serious	40.42	0.88^b (0.79, 0.98)	40.04	0.90 (0.81, 1.00)	45.14	0.71 (0.49, 1.05)
Sexual debut (age in years)
<19	48.54	1.31^c (1.19, 1.44)	48.06	Ref	58.06	Ref
19–29	39.92	Ref	39.53	1.30^c (1.18, 1.43)	46.75	1.51 (0.98, 2.33)
≥30	51.33	1.08 (0.91, 1.28)	48.1	1.07 (0.87, 1.31)	55.59	1.14 (0.81, 1.59)
Most recent HIV test
≤1 year	39.74	Ref	38.82	Ref	50.78	Ref
>1 year	42.70	1.13 (0.98, 1.30)	41.96	1.16 (1.00, 1.35)	48.31	0.83 (0.54, 1.27)
Never	48.94	1.29^c (1.18, 1.42)	48.25	1.32^c (1.20, 1.45)	56.18	1.04 (0.75, 1.44)

CAI: condomless anal intercourse; HIV: human immunodeficiency virus.

p < 0.01.

p < 0.05.

p < 0.001.

A subgroup analysis was conducted due to the statistically significant interaction of age between the frequency of app usage and CAI (p = 0.01). Among those aged <40 years, CAI was marginally associated with frequency of app use. Those using the app 6–10 times per week had a 12% decreased odds of CAI (AOR = 0.88, 95%CI: 0.78–1.00), while there was an 8% decreased odds for those using the app ≥11 times a week (AOR = 0.92, 95%CI: 0.83–1.01). Among those aged ≥40 years old, the frequency of logging onto the app ≥11 times per week was significantly, inversely associated with CAI (AOR = 0.57, 95%CI: 0.40–0.81), but the frequency of logging onto the app 6–10 times per week was not found to be statistically significant (AOR = 0.73, 95%CI: 0.48–1.10) (Table 3).

RCSs incorporated in the multiple logistic regression analyses

Figure 2 shows the association between frequency of logging onto the app and CAI modeled by RCSs and comparison of predicted odds ratio point estimates for CAI. In the study population, the app login frequency was associated with CAI in a curvilinear shape after adjustment of potential confounders, with higher frequency of logging onto the app associated with lower odds of CAI. Similar inverse trends were observed in groups aged <40 years. For those who aged ≥40 years, the relationship between app use frequency and CAI was linear, with higher frequency of app use being associated with lower CAI.

Figure 2.

App use frequency and odds of condomless anal intercourse among (a) the study population, (b) those aged < 40 years, and (c) those aged ≥40 years old. Odds ratio and 95%CI from logistic regression using RCSs with app use frequency equals to 3 times per week as the reference group. Multivariable adjustment in the study population for age, education level, marriage, hukou, time since registering on the app, sexual debut, self-perception of HIV epidemic in Beijing, and HIV testing history at baseline. CAI: condomless anal intercourse.

Discussion

Key findings

We explored the relationship between gay dating app use frequency and CAI among MSM app users in Beijing, China. Although previous studies have examined associations between app use and high-risk sexual behaviors,^17,20 our study expands on the literature by focusing on the frequency of app use in detail. In this study of 9280 MSM gay dating app users, almost half logged onto the app ≥11 times per week, which was consistent with other studies.^20,31,32 Additionally, more than 30% of participants completed the survey, higher than rates reported in other similar studies,^33,34 indicating acceptability and representativeness of the study. MSM are regular and frequent gay dating apps users, with almost twice-per-day use common among this group. As such, gay dating apps can be an effective way to reach key populations and offer health promotion messaging to often hard to target groups.^35–39 These apps could be a private and quick approach to spreading health information to often underserved populations.

In this study, frequency of app use was inversely associated with CAI. This finding indicates positive associations between frequent app use and offline health behaviors, which is supported by other previous research.^18,27,40 Yeo and Ng²⁷ conducted a survey of young men who have sex with men (YMSM) in Hong Kong and found that higher frequency of gay dating app usage reduced the odds of inconsistent condom use with sexual partners after adjusting for self-reported HIV status and other potential covariates. Another study, which was conducted among 18- to 24-year-old Grindr-using MSM in Los Angeles in 2011, reported that the odds of CAI among those who used Grindr five or more times per day was 5% lower than those who used the app less than five times per day.¹⁸ Gay dating apps provide a mobile platform for health information dissemination, for example providing social support⁴¹ and disseminating enriched health evidence.⁴² This may explain the positive association between frequent app use and safe offline behaviors. More importantly, the popularity of the Blued app in China has attracted researchers to develop and conduct effective interventions for app users.^38,43 As such, Blued users who frequently log onto the app likely have a higher probability of receiving these kinds of health-related online interventions.

In our study, men with CAI were more likely to be those aged 40 years or older, which is consistent with a recent meta-analysis reporting a higher pooled prevalence of CAI among older MSM compared with younger MSM.⁴⁴ Certain cultural aspects of Chinese life have encouraged MSM to marry partners of the opposite sex. Confucianism, a traditional philosophy in China, which emphasizes duty, obedience, and filial piety, regards marriage as a family business, with little emphasis on love. These beliefs may prompt older MSM to marry a woman and hide their gay identity in order to protect their family and social status.⁴⁵ With approximately 17% of MSM currently married and 26.3% of MSM having female sexual partners in past six months,²² this group may be more likely to engage in condomless anal sex activities with casual partners in order to meet their need for sex.⁴⁶ Findings from this study also suggest that, for those aged 40 years and older, the relationship between app login frequency and CAI was linear, meaning the higher frequency of app use, the lower odds of CAI. While existing studies support that gay dating apps are dominated by YMSM,²¹ future studies should focus on how to guide social media use and design interventions for older MSM, particularly as lower risk perception among older MSM hinders condom use during anal sex.⁴⁷

Implications

This study showed the optimistic role of gay dating apps in reducing CAI among MSM in China. Future longitudinal studies are needed to evaluate how frequent app use and app-based interventions can impact high-risk sexual behaviors, with additional comparison of app-based data collection against self-reported high-risk behavioral data.

Limitations

Several limitations of this study exist. First, convenience sampling was used to recruit participants, who may not be representative of the general MSM community throughout China. Second approximately 30% of participants completed the study survey. These individuals may be better engaged in care and have access to health-related resources that promote less-risky sexual behaviors compared to the 70% who did not complete the survey. However, more than 9000 individuals participated in the study, which was anonymously administered via a mobile platform, likely decreasing this volunteer bias. Third, survey data were collected from an online, self-reported questionnaire, which may be subject to reporting bias. To address this, we specified in the informed consent that all responses were de-identified in order to promote anonymity and reduce such bias. Fourth, same source bias may be present, as participants self-reported both their frequency of app use and CAI.⁴⁸ Additional follow-up should compare app use frequency reported by participants against data collected by the app. Fifth, a clear, causal effect of frequency of logging onto the app and CAI could not be ascertained from this cross-sectional study. Authors did not consider other potential confounders between the relationship of app use and CAI, for example usage of multiple social networking apps,¹⁶ or drug and alcohol use, which have been found to be associated with CAI among app users.⁴⁹ Finally, only HIV-negative individuals were included in the study, potentially impacting the risk decisions of participants. A follow-up study comparing frequency of app use and CAI in both HIV-negative and HIV-positive MSM would further bolster the important findings of this study.

Conclusions

The results of this study suggest that frequent use of gay dating apps may reduce CAI among MSM, supporting the role of gay dating apps in the field of HIV prevention and control.

Footnotes

Acknowledgements

The authors would like to thank all the study subjects for their voluntary participation in the study and thank Blued staff members for their assistance in conducting the study. The corresponding author had full access to all the data and had final responsibility for the decision to submit for publication.

Authors’ contribution

QL, GM, and ZW designed the study. QL, ZC, YM, GM, XL, JX, KR, and YZ contributed to data collection. QL, GM, and SRS performed the analysis. QL, ZW, GM, and SRS interpreted the results and developed the initial draft. All authors contributed to manuscript revisions and approved the final version for publication.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Disclaimer

The opinions expressed herein reflect the collective views of the co-authors and do not necessarily represent the official position of the National Center for AIDS/STD Control and Prevention, China CDC.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the National Science and Technology Major Project on Prevention and Treatment of Major Infectious Diseases Including AIDS and Viral Hepatitis from the National Health Commission of the People’s Republic of China (2018ZX10721102). The funding organization had no role in the development of study design, collection, analysis, and interpretation of data; the writing of the report; or the final decision to submit the manuscript for publication.

References

Beyrer

Baral

van Griensven

, et al. Global epidemiology of HIV infection in men who have sex with men. Lancet 2012; 380: 367–377.

Chow

Wilson

Zhang

, et al. Human immunodeficiency virus prevalence is increasing among men who have sex with men in China: findings from a review and meta-analysis. Sex Transm Dis 2011; 38: 845–857.

Tang

Meyers

, et al. HIV epidemiology and responses among men who have sex with men and transgender individuals in China: a scoping review. BMC Infect Dis 2016; 16: 588.

National Health and Family Planning Commission of the People’s Republic of China and The Joint United Nations Programme on HIV and AIDS. 2015 China AIDS response progress report. Geneva: UNAIDS, 2015.

NCAIDS., NCSTD and China C. Update on the AIDS/STD epidemic in China in December 2017. Chin J AIDS STD 2018; 24: 111.

Yin

Liu

Zhu

, et al. The prevalence, temporal trends, and geographical distribution of HIV-1 subtypes among men who have sex with men in China: a systematic review and meta-analysis. Epidemiol Infect 2019; 147: e83.

Wang

, et al. Analysis of epidemiological characteristics of HIV/AIDS in Beijing, 2017. Capital J Public Health 2018; 12: 282–284.

National Center for AIDS/STD Control and Prevention and China CDC. 2017 China national HIV/STD/HCV programs implementation report. Beijing: National Center for AIDS/STD Control and Prevention, China CDC, 2017.

Preston

D’Augelli

Kassab

, et al. The relationship of stigma to the sexual risk behavior of rural men who have sex with men. AIDS Educ Prev 2007; 19: 218–230.

10.

Tang

Mao

Tang

, et al. Disclosure of sexual orientation to health professionals in China: results from an online cross-sectional study. J Int AIDS Soc 2017; 20: 21416.

11.

Lau

Lin

Hao

, et al. Public health challenges of the emerging HIV epidemic among men who have sex with men in China. Public Health 2011; 124: 260–265.

12.

Wirtz

Zelaya

Peryshkina

, et al. Social and structural risks for HIV among migrant and immigrant men who have sex with men in Moscow, Russia: implications for prevention. AIDS Care 2014; 26: 387–395.

13.

Grindr. Wikipedia. Available: https://en.wikipedia.org/wiki/Grindr (accessed 1 June 2019).

14.

Chen

Chu

, et al. Using a social networking “app” to establish an open cohort for assessing HIV incidence among men who have sex with men in Beijing, China. In: The 22nd International AIDS Conference Abstract Supplement. Amsterdam: Journal of the International AIDS Society 2018, 21(S6): e25148 (p. 166).

15.

Muessig

Nekkanti

Bauermeister

, et al. A systematic review of recent smartphone, internet and web 2.0 interventions to address the HIV continuum of care. Curr HIV/AIDS Rep 2015; 12: 173–190.

16.

Holloway

Rice

Gibbs

, et al. Acceptability of smartphone application-based HIV prevention among young men who have sex with men. AIDS Behav 2014; 18: 285–296.

17.

Bien

Best

Muessig

, et al. Gay apps for seeking sex partners in China: implications for MSM sexual health. AIDS Behav 2015; 19: 941–946.

18.

Winetrobe

Rice

Bauermeister

, et al. Associations of unprotected anal intercourse with Grindr-met partners among Grindr-using young men who have sex with men in Los Angeles. AIDS Care 2014; 26: 1303–1308.

19.

Holloway

Pulsipher

Gibbs

, et al. Network influences on the sexual risk behaviors of gay, bisexual and other men who have sex with men using geosocial networking applications. AIDS Behav 2015; 19: 112–122.

20.

Lehmiller

Ioerger

Social networking smartphone applications and sexual health outcomes among men who have sex with men. PloS One 2014; 9: e86603.

21.

Zou

Fan

Characteristics of men who have sex with men who use smartphone geosocial networking applications and implications for HIV interventions: a systematic review and meta-analysis. Arch Sex Behav 2017; 46: 885–894.

22.

Chow

Wilson

Zhang

What is the potential for bisexual men in China to act as a bridge of HIV transmission to the female population? Behavioural evidence from a systematic review and meta-analysis. BMC Infect Dis 2011; 11: 242.

23.

Blued. 2015 white paper on the big data of China’s gay community. Beijing: January 11, 2016. Available: http://www.199it.com/archives/428104.html. (accessed 1 June 2019)

24.

Broder

McGrath

Cisneros

Questionnaire development: face validity and item impact testing of the child oral health impact profile. Community Dent Oral Epidemiol 2007; 35: 8–19.

25.

Chen

Internal migration and health: re-examining the healthy migrant phenomenon in China. Soc Sci Med 2011; 72: 1294–1301.

26.

Mor

Davidovich

Same-sex sexual attraction, behavior, and practices of Jewish men in Israel and the association with HIV prevalence. AIDS Care 2016; 28: 64–67.

27.

Yeo

YL.

Sexual risk behaviors among apps-using young men who have sex with men in Hong Kong. AIDS Care 2016; 28: 314–318.

28.

Gaydos

Jett-Goheen

Barnes

, et al. Use of a risk quiz to predict infection for sexually transmitted infections: a retrospective analysis of acceptability and positivity. Sex Transm Infect 2016; 92: 44–48.

29.

Greenland

Pearl

Robins

JM.

Causal diagrams for epidemiologic research. Epidemiology 1999; 10: 37–48.

30.

Desquilbet

Mariotti

Dose-response analyses using restricted cubic spline functions in public health research. Stat Med 2010; 29: 1037–1057.

31.

Badal

Stryker

DeLuca

, et al. Swipe right: dating website and app use among men who have sex with men. AIDS Behav 2018; 22: 1265–1272.

32.

Lorimer

Flowers

Davis

, et al. Young men who have sex with men’s use of social and sexual media and sex-risk associations: cross-sectional, online survey across four countries. Sex Transm Infect 2016; 92: 371–376.

33.

Rosser

Oakes

Konstan

, et al. Reducing HIV risk behavior of men who have sex with men through persuasive computing: results of the men’s INTernet study-II. AIDS 2010; 24: 2099–2107.

34.

Carpenter

Stoner

Mikko

, et al. Efficacy of a web-based intervention to reduce sexual risk in men who have sex with men. AIDS Behav 2010; 14: 549–557.

35.

Lelutiu-Weinberger

Pachankis

Gamarel

, et al. Feasibility, acceptability, and preliminary efficacy of a live-chat social media intervention to reduce HIV risk among young men who have sex with men. AIDS Behav 2015; 19: 1214–1227.

36.

Mustanski

Garofalo

Monahan

, et al. Feasibility, acceptability, and preliminary efficacy of an online HIV prevention program for diverse young men who have sex with men: the keep it up! intervention. AIDS Behav 2013; 17: 2999–3012.

37.

Young

Cumberland

Lee

, et al. Social networking technologies as an emerging tool for HIV prevention: a cluster randomized trial. Ann Intern Med 2013; 159: 318–324.

38.

Tang

Han

Best

, et al. Crowdsourcing HIV test promotion videos: a noninferiority randomized controlled trial in China. Clin Infect Dis 2016; 62: 1436–1442.

39.

Christensen

Miller

Appleby

, et al. Reducing shame in a game that predicts HIV risk reduction for young adult MSM: a randomized trial delivered nationally over the web. J Int AIDS Soc 2013; 16: 18716.

40.

Cao

Liu

Durvasula

, et al. Social media engagement and HIV testing among men who have sex with men in China: a nationwide cross-sectional survey. J Med Internet Res 2017; 19: e251.

41.

Han

, et al. Weibo friends with benefits for people live with HIV/AIDS? The implications of Weibo use for enacted social support, perceived social support and health outcomes. Soc Sci Med 2018; 211: 157–163.

42.

Taggart

Grewe

Conserve

, et al. Social media and HIV: a systematic review of uses of social media in HIV communication. J Med Internet Res 2015; 17: e248.

43.

Tang

Wei

Cao

, et al. Crowdsourcing to expand HIV testing among men who have sex with men in China: a closed cohort stepped wedge cluster randomized controlled trial. PLoS Med 2018; 15: e1002645.

44.

Qian

, et al. High prevalence of HIV infection and unprotected anal intercourse among older men who have sex with men in China: a systematic review and meta-analysis. BMC Infect Dis 2014; 531.

45.

Zhang

Liu

, et al. AIDS-related high-risk behaviors and affecting factors of men who have sex with men (MSM) in mainland China. Chin J Sex Transm Inf 2001; 1: 7–16.

46.

Zhang

KC.

Travel on the border of the implicit marriage and the institutional marriage: analysis on the marriage models among Chinese gay population. J South China Normal Univ 2013; 6: 22–30.

47.

Zhang

Chen

ZY.

Qualitative study on unprotected sexual behaviors among male students who have sex with men in Beijing. Chin J Prev Med 2018; 65: 1234–1238.

48.

Gullón

Bilal

Franco

Physical activity environment measurement and same source bias. Gac Sanit 2014; 28: 344–345.

49.

White

Finneran

Sato

, et al. Sex, HIV, and the internet: exploring variations in the online profiles of MSM in the United States. Am J Mens Health 2014; 8: 289–299.