JMIR Publications

ABSTRACT

Background:

Regular HIV/STI testing for men who have sex with men (MSM) is an important means of infection prevention, the adoption of which in the community remains suboptimal.

Objective:

This study aims to pilot “facilitated testing” of HIV/STI through a designated web system integrated with self-sampling to enhance the engagement of MSM for regular testing.

Methods:

This is a one-year cohort study on HIV-negative MSM aged 18 years or above. Through a web-based system, participants made appointments for baseline and followup visits at 3-monthly intervals. On-site blood sampling was performed for HIV and syphilis tests, alongside self-collection of pharyngeal swabs, rectal swabs, and urine samples for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) testing. Self-sampling and self-testing kits were delivered as alternatives to on-site visits in followups. Baseline, monthly followup and exit surveys were conducted for collecting sexual behaviour, STI/HIV testing history, and post-self-sampling scoring. Full engagement as defined by having made at least 3 visits in over half a year was compared with partial engagement in the bivariable logistic regression model. Factors associated with incident STI were examined in the multivariable cox regression model.

Results:

Between August 2019 and October 2020, 204 MSM were recruited. A majority (93%) were Chinese, with a median age of 31 years old (IQR=26-39), and 58% were PrEP-experienced at baseline. Some 146 MSM were in full engagement (72%), which was associated with experiencing incident STI during followups (OR=4.23, 95%C.I.=1.63-10.94), seeking medical referral after STI detection (OR=10.25, 95%C.I.=3.25-29.79), and synchronised schedule of HIV/STI testing with PrEP visits (OR=51.85, 95%C.I.=19.30-139.34). No incident HIV was detected in the followup period. At baseline, the overall STI (CT, NG or syphilis) prevalence was 30% (95%C.I.=24%-36%), with CT at 18% (95%C.I.=12%-23%), NG at 13% (95%C.I.=8%-17%), and syphilis at 5% (95%C.I.=2%-8%). In followup, the incidence of any STI was 59.08/100 person-years (py), CT was 33.05/100py, NG was 29.86/100py, and syphilis was 10.4/100py. History of group sex (adjusted hazard ratio(aHR)=3.00, P=.04), and baseline condom usage level during chemsex engagement (aHR for never using a condom=2.15, P=.02; aHR for using a condom for less than half of sex acts=2.35, P=.007; no chemsex engagement as reference) were the main predictors of any incident STI. The detection rates of CT and NG in urine samples were lower than pharyngeal swabs and rectal swabs. Their scores for convenience, confidence of correct sampling, and accuracy of self-sampling were high (7 to 8 out of 10).

Conclusions:

Both baseline prevalence and incidence of STI were high among MSM engaged in regular testing. Their degree of engagement in regular STI/HIV testing was associated with profile of sexual behaviours and history of health-seeking behaviours. Self-sampling for HIV/STI testing could potentially supplement conventional method in the expansion of testing coverage.