Predictors of antiretroviral therapy initiation: a cross-sectional study among Chinese HIV-infected men who have sex with men

Background Early antiretroviral therapy (ART) initiation is crucial to achieve HIV viral suppression and reduce transmission. HIV-infected Chinese men who have sex with men (MSM) were less likely to initiate ART than other HIV-infected individuals. We assessed predictors of ART initiation among Chinese MSM. Methods In 2010–2011, a cross-sectional study was conducted among MSM in Beijing, China. We examined ART initiation within the subgroup who were diagnosed with HIV infection prior to participation in the survey. Logistic regression models were fitted to evaluate socio-demographic and behavioral factors associated with ART initiation. The eligibility criterion in the 2010/2011 national HIV treatment guidelines was CD4 cell count <350 cells/μL or World Health Organization (WHO) clinical stage III/IV. Results Of 238 eligible HIV-infected participants, the median duration of HIV infection was 15 months (range: 31 days-12 years); 62 (26.1 %) had initiated ART. Among 103 men with CD4 counts <350 cells/μL, 38 (36.9 %) initiated ART. Being married to a woman (adjusted odd ratios [aOR]: 2.50; 95 % confidence interval [CI]: 1.07-5.87), longer duration of HIV infection (aOR: 10.71; 95 % CI: 3.66-31.32), and syphilis co-infection (aOR: 2.58; 95 % CI: 1.04-6.37) were associated with a higher likelihood of ART initiation. Of 135 men with CD4 count ≥350 cells/μL, 24 (18 %) initiated ART. Being married to a woman (aOR: 4.21; 95 % CI: 1.60-11.06), longer duration of HIV infection (aOR: 22.4; 95 % CI: 2.79-180), older age (aOR: 1.26; 95 % CI: 1.1-1.44), Beijing Hukou (aOR: 4.93; 95 % CI: 1.25-19.33), presence of AIDS-like clinical symptoms (aOR: 3.97; 95 % CI: 1.32-14.0), and history of sexually transmitted infections (aOR: 4.93; 95 % CI: 1.25-19.43) were associated with ART initiation. Compared with men who did not initiated ART, those with ART were more likely to receive counseling on benefits of ART (96.8 % vs. 66.4 %, P = 0 < 0.01), HIV stigma coping strategy (75.8 % vs. 65.9 %, P = 0.04), mental health (66.1 % vs. 52.9 %, P = 0.02), and substance use (46.7 % vs. 36.6 %, P = 0.04). Conclusions We documented low rates of ART initiation among Chinese MSM. Policy changes for expanding ART eligibility and interventions to improve the continuum of HIV care are in progress in China. Impact evaluations can help assess continuing barriers to ART initiation among MSM.


Background
Antiretroviral therapy (ART) use among HIV-infected individuals is crucial for long-term success in reducing HIV transmission and disease progression [1,2] Early initiation of ART can suppress viral replication and reduce harmful immune activation, slowing pathogenesis, and stem the spread of the virus ('treatment-as-prevention') [3,4]. Mathematical modeling and observational studies suggested that the regions with higher ART coverage had reduced HIV transmission between regular and casual sexual partners, lower community viral load, and decreased HIV incidence [5][6][7][8]. The randomized clinical trial HPTN052 confirmed that early initiation of ART could significantly reduce sexual transmission of HIV among serodiscordant couples [9].
Despite these benefits of modern ART [2,8,10], many HIV-infected individuals in China are still not on ART [11]. A 2011 study showed that only 41 % (88,185/ 214,714) HIV-infected individuals had initiated ART [12], though the rate has been increasing steadily in the recent years. Chinese men who have sex with men (MSM) are less likely to start ART compared with other risk groups [13].
HIV epidemic among Chinese MSM is the most robust in all high risk groups [14][15][16]. According to the latest national estimates, HIV prevalence among Chinese MSM uniquely represented a marked uptrend (7.3 % in 2013 vs. 0.9 % in 2003), while in other risk groups it either dropped or maintained at a low level [17]. Maleto-male transmission accounted for 21.4 % of all new infections in 2013, with the fastest increasing rate among all risk groups [17]. As Chinese MSM may marry a women due to social and family pressures [18][19][20], HIVinfected MSM are also likely transmit the virus to their female sexual partner. Therefore, it is a crucial strategy for HIV prevention to expand ART coverage among Chinese MSM. However, there is little knowledge about ART initiation among Chinese MSM. A qualitative study suggested that Chinese MSM may not seek HIV care or initiate ART due to psychological, social, cultural and institutional barriers [21]. We conducted the first quantitative assessment of ART initiation and its predictors among Chinese HIV-infected MSM.

Study design and population
In 2010-2011, we conducted a cross-sectional study among Chinese MSM with a primary objective of assessing the association between male circumcision and HIV risk among MSM in Beijing, China. This study was described elsewhere [22][23][24]. In brief, participants were recruited from HIV/sexually transmitted disease (STD) clinics or the community, and through peer-referral; participants who were tested HIV positive from other studies in Beijing were also recruited. Eligibility criteria included: being a male, living in Beijing, self-reporting ever having sex with men, willingness to provide blood samples for HIV and syphilis serological tests and anal swab specimens for HPV testing, and willingness to provide written informed consent were eligible to participate in the study. All eligible participants were asked to complete a questionnaire interview, undertake genital examinations, and provide blood samples [24]. The study protocol was approved by the Institutional Review Boards of the National Center for AIDS/STD Control and Prevention of the Chinese Center for Disease Control and Prevention, and the Vanderbilt University School of Medicine.

Data collection
Each participant completed a questionnaire on sociodemographic characteristics: age, ethnicity, marital status, education, occupation, registered household in Beijing (or Beijing Hukou), duration of living in Beijing, and sexual orientation. The questionnaire also assessed behavioral risk factors: unprotected insertive or receptive anal sex, most recent CD4+ T-lymphocyte cell count (CD4+ count), and history of sexually transmitted diseases (STD, including chlamydia, gonorrhea, syphilis, herpes simplex type 2, etc.). Those who had been diagnosed with HIV infection prior to participation in this study were also asked about AIDS-like clinical symptoms, experiences of using ART and HIV care and counseling, such as substance use, mental health, stigma coping strategies, benefits of ART, and prevention of HIV transmission.

Statistical analysis
A significant portion of participants had been previously diagnosed with HIV infection, and were asked for ART use. Therefore, we performed a subgroup analysis among these participants to assess their ART initiation and its predictor variables.
We expressed descriptive statistics as proportions for categorical variables and as means and standard deviations, or medians and interquartile ranges for continuous variables, depending on the distribution of the variables. The socio-demographic, clinical, and behavioral characteristics were compared between those who had initiated ART and who had not in the whole study sample and in subgroups by CD4 count of 350 cells/uL, which was the cut point of ART eligibility in Chinese national guidelines during 2010 and 2011.We also examined WHO HIV clinical staging (III or IV) for dividing the subgroups. Continuous variables were compared with Wilcoxon ranksum tests, and categorical variables were compared with the Chi-square or Fisher's exact tests. Proportions of MSM receiving HIV care counseling services between men who did and did not initiate ART were compared using Chisquare tests.
To further assess the strength and direction of the associations between selected predictors and ART initiation stratified by CD4+ count, variables moderately associated with ART initiation (P < 0.2) in bivariable analyses were selected for multivariable logistic regression analyses. In the multivariable logistic regression models, variables with a P < 0.2 in the bivariable logistic regression analyses were preliminarily fitted into each model, respectively, then a manual backward selection procedure was used to conclude the final multivariable models that only retained covariates significantly associated with the outcome(P < 0.05). However, factors considered as a priori predictors or confounders based on prior evidence were forced to retain in the final models regardless of P-value [25]. All statistical analyses were performed using Stata 12.0™ (StataCorp LP, College Station, Texas, USA).

Characteristics of Study Population
Among 1155 participants in the study, 290 were confirmed to be HIV positive (25.1 %), including 48 men newly diagnosed upon enrolled in the current study and 242 previously diagnosed. The 48 newly diagnosed HIV cases were excluded in the analysis since they had had no opportunity to initiate ART upon the survey. Of the 242 previously diagnosed HIV-infected MSM, 4 had missing data on ART initiation, and therefore, 238 (98.3 %) were included in this analysis.
Over a quarter (26.1 %) men had started ART. Compared with those who had not initiated ART, men who had initiated ART were more likely (P < 0.05) to be older, be married to a woman, have Beijing Hukou, live longer in Beijing, have a longer duration of HIV-infection, experience adverse clinical symptoms, have a history of STD, and be co-infected with syphilis. However, those who had initiated ART were less likely (P < 0.05) to be engaged in unprotected insertive or receptive anal sex (Table 1).
Among 103 MSM with CD4 + cell count (CD4 count) <350 cells/μL who met the eligibility criteria for Chinese free ART program during this survey, 36.7 % initiated ART. Those who had initiated ART were more likely (P < 0.05) to be married to a woman, have longer duration of HIV-infection, and have syphilis coinfection, but were less likely (P < 0.05) to have unprotected insertive or receptive anal sex ( Table 2).
Of 135 MSM with CD4 count ≥350 cells/μL who did not meet the criteria, 17.8 % initiated ART. Those who had initiated ART were more likely (P < 0.05) to be older, married to a woman, have Beijing Hukou, live longer in Beijing, have longer duration of HIV-infection, have ever had clinical symptoms, have a history of STD, and have syphilis co-infection, but were less likely (P < 0.05) to have unprotected insertive or receptive anal sex (Table 2).
Factors associated with ART initiation by CD4 count level Table 3 shows the variables associated with ART initiation in both bivariable and multivariable logistic regression

Discussion
To our knowledge, this is the first study on predictors of ART initiation and receipt of HIV care among Chinese HIV-infected MSM. In the study, we found older age and ever having adverse clinical symptoms were positively associated with ART initiation, and this is consistent with findings from other regions [26]. Due to social and family pressure, some Chinese MSM are married with women or have female sexual partners [27]. Our study showed that married MSM were more likely to start ART than unmarried MSM. Married men might have more sense of responsibility for the health of themselves and their wives, and therefore have started ART; however, more research is needed to explore the reasons. Beijing Hukou and longer duration of living in Beijing were also positively associated with ART initiation. Migrant MSM, particularly recent migrants, might not know that free ART is for everyone, or where they could get it. It is suggested HIV intervention programs should give special emphasis among migrant MSM who account for the majority of MSM population in many large cities. In addition, Men who had initiated ART reported receiving more HIV care counseling services than those without ART, and therefore could get more health benefits from the counseling. Our findings suggest considerations for the design of future intervention programs. The positive association Ever had at least one of these adverse clinical symptoms in the past 6 months: severe weight loss (>10 %), fever (>1 month, continuing or intermittent), chronic diarrhea (>1 month), severe bacterial infection (e.g. pneumonia), oral candidiasis, oral leukoplakia, tuberculosis other stage III or stage IV symptoms between being married and ART initiation observed among both ART eligible and ineligible MSM (using criteria in place in 2010-2011) suggest that a built-in educational or psychological counseling session would be necessary for cultivating the sense of responsibility for spouses and family among married HIV-infected MSM, which might facilitate early ART initiation. Given that the longer time from HIV diagnosis to the survey and older age was positively associated with ART initiation, greater government and community efforts to target and engage men earlier and more intensively may be effective, alongside linking recently infected MSM to HIV care for ART initiation. The benefits of legal residence (Beijing Hukou) in accessing HIV care reflect both one's ability to use local medical services and social benefits, and also one's motivation and willingness in seeking these services. People without local residence might be more financially, physically, and/or emotionally burdened as compared to local residents. In our study, having Beijing Hukou played a crucial role in ART initiation among MSM who did not meet free ART criterion. In the meantime, those who had longer duration living in Beijing demonstrated a higher likelihood of proactive ART initiation. Thus, our findings suggest the need for researching the sexual network and cooperating with local gay organizations with outreach to the migrant MSM who recently move to Beijing and/or are transiently working in Beijing. China started its national free ART program in 2003 [28]. This program has significantly reduced mortality among HIV-infected individuals [29,30]. However, our study showed that only 26 % HIV-infected participants who had been infected with HIV for a median 15 months (IQR, 6-31) initiated ART, and the rate was lower than other risk groups in China [31][32][33] and MSM in other countries [34][35][36][37][38]. Men who met treatment criteria (36.7 %) were more likely to initiated ART than those who did not (17.8 %), but both were quite low. Our study examined ART initiation among Chinese for the first time. Previous surveys suggested considerable willingness to receive ART [13,39], yet this willingness seems not translating into uptake of ART, where studies revealed low ART use [14,[40][41][42]. There are multiple reasons for low ART use. First, Chinese free ART program were originally developed to provide "Four Free and One Care" (free ARV drugs, free prevention of mother-to-child transmission, free voluntary counseling and testing, free schooling for children orphaned by AIDS, and care to people living with HIV/AIDS) to HIV-infected former plasma donors and persons who injected drugs and their family members. Chinese government has vigorously promoted ART use in these populations [43]. In contrast, the promotion activities of ART use among HIV-infected MSM were sporadic prior to 2011, mainly through programs funded by international agencies such as Global Fund for AIDS and Gates Foundation. HIV-infected MSM might not be aware that they can also use the Chinese free ART program. Second, the HIV epidemic among Chinese MSM is more recent compared with those in other risk groups. Although prevention-oriented intervention programs exist for Chinese MSM, ART is relatively new and therefore unfamiliar to many [44]. Third, the majority of HIV-infected MSM are generally young (<40 years), well educated, and living in large cities, so they are well accessible to HIV-related knowledge and information. They are concerned about side-effects of first-line generic ARVs in the Chinese free ART program and possible drug resistance due to failure in treatment adherence; they are concerned about no alternative drugs due to the limited number of available antiretroviral drugs (ARVs) in Chinese ART program [45][46][47]. Fourth, Chinese MSM face dual stigma of HIV and homosexuality, and they may encounter discrimination from the community and healthcare providers, which prevents them from seeking HIV care and treatment [48][49][50]. Last, the stricter eligibility criteria in earlier years might also exclude some MSM from treatment. China ART program relaxed its criteria to CD4 count <500 cells/μL in 2013. Actually in some cities, HIV-infected individuals can start ART immediately at diagnosis, like do in high income countries [51][52][53].
We recognized the limitations of our study: (1) With a relatively small sample of HIV-infected Chinese MSM living in Beijing, findings may not represent very precise estimates or be generalizable to other settings; (2) Data on socio-demographic characteristics, risky behaviors, and some clinical factors were based on self-reporting, subjecting these data to recall and social desirability biases; (3) In this cross-sectional study, we were not able to ascertain some key information, such as the exact time of ART initiation, and viral load or other markers for viral replication or immune activation at HIV diagnosis or ART initiation. These factors might also affect the physician's decision on prescribing ART. However, the main criteria for ART initiation are based on CD4 count and WHO clinical staging, but not on HIV viral load according to Chinese ART protocol.

Conclusions
In conclusion, the rates of ART initiation among Chinese MSM were low as of 2011. Compared to other subgroups, HIV-infected Chinese MSM are less likely to initiate ART. This is the first study in China to investigating factors associated with ART initiation in this marginalized population. Policy changes for expanding ART eligibility and interventions to improve the continuum of HIV care are in progress in China. Impact evaluations can help assess continuing barriers to ART initiation among MSM.