Persons with sexually transmitted infections (STIs) are prime PrEP candidates, but the accuracy of using self-report to determine PrEP use is unknown. We used a novel approach to objectively measure PrEP use in a sentinel population largely representative of New York City (NYC) residents diagnosed with STIs, providing information about PrEP uptake and the accuracy of self-report.
We systematically accrued remnant serum samples from the following HIV-negative patients screened for syphilis at NYC Sexual Health Clinics (January-June 2019): men who have sex with men (MSM) and women with chlamydia (CT), gonorrhea (GC), and /or early syphilis (ES). Samples were tested for tenofovir/emtricitabine (detected vs. not detected) using a validated liquid chromatography-mass spectrometry (LC-MS) assay. Pairing test results with medical records, we assessed: 1) PrEP use on the day of STI diagnosis, 2) agreement of LC-MS assay with self-reported PrEP, and 3) correlates of PrEP use among MSM with rectal CT/GC or ES.
PrEP use among 744 patients (331 MSM with rectal CT/GC, 122 MSM with ES, 85 MSM with urethral GC, 109 MSM with co-infections, 97 women with GC/ES) was 32.8% (95% CI, 29.4%-36.3%). PrEP use was highest among White patients (45%) and MSM with ES (44%), and lowest among Black patients (20%) and women (2%). Agreement between LC-MS and self-reported PrEP use was 91%. Among MSM with rectal CT/GC or ES, PrEP use was associated with age [adjusted prevalence ratio (aPR)=1.6 (95% CI, 1.0-2.5) for ages 25-34 and aPR=2.0 (1.2-3.4) for ages 35-44, vs. 15-24 years]; number of recent sex partners [aPR=2.1 (1.3-3.4) for 6-10 partners and aPR=2.0 (1.2-3.3) for >10 partners, vs. <2 partners]; having sex/needle-sharing partners with HIV [aPR=1.4 (1.0-1.8)]; and inconsistently vs. always using condoms [aPR=3.1 (1.5-6.3)]. Race/ethnicity and past-year history of CT/GC/ES diagnoses or post-exposure prophylaxis were not associated.
This is the first study to analyze routinely collected remnant samples from STI clinics for a PrEP biomarker. Although the accuracy of self-reported PrEP was high, only 1 in 3 people with a newly diagnosed STI was on PrEP. PrEP use was associated with most measured HIV risk factors, but it is critical to increase use in racial minority populations and women. Surveillance studies using remnant samples can assess the accuracy of self-reported PrEP use in other settings and evaluate the success of interventions to increase PrEP uptake in high-risk populations.