Abstract
The rate of syphilis infections among sexual minority men (SMM) has continued to increase in recent decades. As such, this analysis sought to identify demographic, biological, and behavioral factors associated with recent syphilis infection in emerging adult SMM. Data were drawn from a 3-year cohort study of emerging adult SMM (n = 665), from July 2014 to March 2019. Biannual study assessments included rapid HIV testing and behavioral surveys. At baseline, and at the 18- and 36-month time points, participants underwent chlamydia, gonorrhea, and syphilis screening. Generalized estimating equations were used to generate four models of repeated syphilis screening. In this racially/ethnically and socioeconomically diverse sample of SMM, 5.0% of participants tested positive for syphilis at baseline and 9.0% had an infection at the subsequent time points. Across all models, racial/ethnic minority SMM had higher odds of syphilis. Higher odds of syphilis infection were also significantly associated with more frequent condomless anal sex, more frequent marijuana use, HIV seropositivity, not currently using pre-exposure prophylaxis (PrEP), and not receiving syphilis testing in the previous 6 months; lower odds were associated with more frequent oral sex and more frequent alcohol use. These findings support current screening guidelines based on SMM who may be at increased risk for sexually transmitted infection (STI) acquisition, such as people living with HIV or those who engage in condomless sex. Further, our findings of reduced syphilis incidence among those who are on PrEP and engaged in regular STI testing support existing efforts to increase the availability and accessibility of preventive sexual health care for SMM.
Keywords: syphilis, pre-exposure prophylaxis, sexual minority men, screening
Introduction
Incidence of syphilis, a sexually transmitted infection (STI) caused by the bacteria Treponema pallidum, has increased substantially over the past decade in the United States.1 Syphilis elimination efforts dwindled in 2000–2001 as reported syphilis cases reached a historic low.2 Subsequently, incident cases of primary and secondary (P&S) syphilis have increased each year, with an estimated increase of 175% from 2008 to 2018, posing a major public health threat.1 This trend has largely been attributed to gay, bisexual, and other men who have sex with men [sexual minority men (SMM)], who bear disproportionately higher burdens of syphilis compared with the general population.3,4
In 2019, men accounted for ∼83% of all new syphilis infections in the United States, 56.7% of whom were SMM.3 These disparities are even more stark in regions such as New York City (NYC), where 96.2% of new P&S syphilis cases in 2016 were among men and 87.7% of those men reported sex with male partners.4 Further, recent increases in syphilis cases have been attributed to a greater proportion of “repeaters,” or individuals who have had syphilis in the past and become reinfected.5
This increased syphilis risk among SMM may be explained by several biopsychosocial and behavioral risk factors. Syndemic factors most commonly linked to overall STI risk among SMM include substance use, condomless sexual behaviors, and mental health burden.6,7 SMM are more likely than their heterosexual counterparts to report sexual behaviors associated with greater risk for STI acquisition, including condomless sex, earlier age of sexual debut, greater number of sexual partners, and casual or nonmonogamous relationships.8–11 Research has similarly consistently documented that alcohol use and use of other drugs [i.e., cocaine, methylenedioxymethamphetamine (MDMA), gamma hydroxybutyrate (GHB), ketamine, and methamphetamine—commonly referred to as club drug use, sexualized drug use, or chem-sex] is commonly associated with increased incidence and prevalence of STIs, including syphilis.12–16
In P&S syphilis cases, racial/ethnic disparities among SMM are documented; 64.8% of cases among SMM in 2019 were men of color, suggesting that interpersonal or structural racism may play a role.3 Previous research has suggested that such racial inequities are not due to increased engagement in transmission/acquisition behaviors, but rather are explained by social and structural drivers of risk, such as sexual networks with higher prevalence of HIV/STIs.17
Compared with all other populations, SMM face the highest lifetime risk of HIV acquisition.18 Recent efforts to reduce HIV incidence have focused on biomedical prevention strategies such as pre-exposure prophylaxis (PrEP). Clinical guidelines for oral PrEP care require SMM to have a prescribing provider and undergo quarterly visits for blood work, including HIV/STI testing.19,20 While this screening likely explains the inconsistently documented increased STI rates among PrEP users, it also enables patients to receive proper and more timely treatment.21–23
Underlying all these factors are the social and structural contexts that influence policy, impact level of engagement and quality of health care interactions, and may deleteriously affect the health and well-being of SMM. These include stigma and bias related to sexual behavior, sexual orientation, gender, and race.24,25
Drawing on data from a longitudinal study of emerging adult SMM in NYC, the present analysis examines behavioral (e.g., substance use, sexual behavior, health care utilization), biological (e.g., HIV status, co-occurring STI diagnosis), and sociodemographic factors associated with syphilis diagnosis in this population.
Methods
Study design
The P18 Cohort Study was a longitudinal investigation of syndemic conditions, focused on HIV and other STIs, sexual behavior, substance use, and mental health burden among SMM; study methods have been previously published.6 Briefly, participants were recruited between 2014 and 2016 and completed biannual study visits over the course of 3 years. At baseline, participants were eligible if they were 22–23 years old, were assigned male at birth, self-reported an HIV-negative status, self-reported having sex with a man in the previous 6 months, and lived in the NYC metropolitan area. At each study visit, participants completed an Audio Computer-Assisted Self-Interview (ACASI), an interviewer-administered Timeline Follow Back (TLFB) assessing the previous 30 days of substance use and sexual behavior, and a rapid HIV test with pre- and post-test counseling.
At baseline, and at 18 and 36 months, participants also underwent multi-site STI screening. Participants provided written informed consent for participation and all study activities were approved by the New York University and Rutgers University Biomedical and Health Sciences Institutional Review Boards.
Measures
Sociodemographic characteristics
Participants self-reported sociodemographic characteristics as part of the ACASI. Race/ethnicity was collapsed into a single variable as Hispanic/Latine, Black, White, Asian/Pacific Islander, multiracial, and Native American/Alaska Native or another race/ethnicity.
Sexual behavior and substance use
Details of the TLFB measure have been published elsewhere.6 Briefly, participants reported event-level data on their substance use for each of the 30 days before their study visit; substances included alcohol, marijuana, club drugs [methamphetamine, ecstasy (MDMA), cocaine, ketamine, and GHB], and prescription medications that were not prescribed or were used other than as prescribed. In addition, participants reported the days on which they engaged in oral, vaginal, or anal sex; whether or not a condom or another barrier was used for each; and whether each activity was insertive or receptive. Each partner was denoted with a different letter to ascertain the total number of unique partners in the 30-day period. In addition, we created variables for the total days of each sexual behavior without condoms/barriers (e.g., days of condomless receptive anal sex) and for the total days of using each substance or group of substances (i.e., alcohol, marijuana, club drugs).
Health care utilization
As part of the ACASI, participants reported on a number of factors regarding their health care utilization, including whether they are currently using PrEP, whether they have a current primary care provider, and whether they have been tested for syphilis in the previous 6 months, all of which were recoded as binary variables.
HIV and STI status
At each study visit, participants were tested for HIV using the Alere Determine rapid HIV-1/2 Ag/Ab combination test, unless they previously tested positive for HIV or were deemed unable to be tested by trained study staff during pretest counseling. Preliminary positive rapid test results were verified by confirmatory RNA testing.
At baseline, and at the 18- and 36-month visits, participants also underwent screening for syphilis, chlamydia, and gonorrhea. Screening for pharyngeal, rectal, and urethral chlamydia and gonorrhea infection was conducted using oral swabs, self-administered anal swabs, and urine samples, respectively. A single, binary variable was coded to measure whether a participant tested positive for an STI other than syphilis.
Syphilis screening was performed using venous blood samples. Samples were subjected to a reverse algorithm for syphilis screening. Blood samples were first screened for treponemal-specific antibodies using an enzyme-linked immunosorbent assay (ELISA). Reactive or equivocal ELISA results automatically reflexed to rapid plasma reagin (RPR) testing. When there was discrepancy between immune assay and RPR reactivity, samples were subjected to a fluorescent treponemal antibody-absorption assay for confirmation. For this analysis, RPR results were used as the primary outcome. Due to the poor ability of treponemal-specific antibody assays to differentiate recent and remote infection, or whether infection has been previously treated, the RPR titer is a more reliable indicator of recent or active infection, and thus more appropriate for this longitudinal design.26
Analytic plan
Variable selection was informed by a syndemic framework and known syphilis risk factors such as condomless sex.7 The additional inclusion of health care utilization variables was intended to standardize screening practices, which has been suggested to reduce ascertainment bias.27 Descriptive statistics assessed the distribution of variables of interest for the entire sample. Differences based on RPR results were then examined for the total sample and across the three time points (baseline, and 18 and 36 months) using Pearson's chi-square test for categorical variables and one-way ANOVA for continuous variables. We then used generalized estimating equations (GEE) to create binary logistic models with an independence working correlation structure and RPR reactive versus RPR nonreactive as the dichotomous outcome. Given the well-documented racial disparities in syphilis prevalence and the plausible associations between race and other factors of interest, each model controlled for race/ethnicity.
Four models were constructed to account for the following: sexual behavior within the 30 days before assessment (total days of oral sex received, oral sex given, condomless anal sex insertive, condomless anal sex receptive, and number of unique partners); substance use within the 30 days before assessment (total days of club drug use, alcohol use, and marijuana use); HIV and other STI status at the time of assessment (positive for gonorrhea and/or chlamydia at any site, and positive for HIV); and health care utilization (current PrEP use, current primary care provider, and accessing syphilis testing in the previous 6 months). The final health care model also controlled for HIV status to avoid potential confounding. In addition to race variables, all variables were entered simultaneously in each model. All analyses were conducted using IBM SPSS Statistics for Windows, version 28 (IBM Corp., Armonk, NY) and significance was set at p < 0.05.
Results
Baseline sample characteristics
The study enrolled a racially/ethnically diverse sample of SMM (n = 665; Table 1). At baseline, a minority of participants tested positive for HIV or an STI other than syphilis and few participants reported current PrEP use. About two-thirds of the sample reported being tested for syphilis in the previous 6 months, and less than half reported having a current primary care provider. There were no significant differences in attrition across the three study time points based on demographics, sexual behavior, or substance use, although some variables naturally changed significantly across the 3-year period (e.g., increased HIV incidence, increased PrEP uptake).
Table 1.
Baseline Sample Characteristics and Bivariate Associations with Reactive Syphilis Test at Any Time Point (n = 665)
| Baseline sample, % (n) | Total RPR+, % (n) | Total RPR−, % (n) | χ2 (p)a | |
|---|---|---|---|---|
| Race/ethnicity | ||||
| Hispanic/Latine | 31.9 (212) | 8.0 (38) | 92.0 (437) | 32.33 (<0.001) |
| Black | 25.6 (170) | 10.2 (42) | 89.8 (371) | |
| Asian/Pacific Islander | 7.4 (49) | 3.5 (4) | 96.5 (110) | |
| Multiracial | 6.5 (43) | 7.9 (8) | 92.1 (93) | |
| Native American/Alaska Native/other | 3.5 (23) | 18.0 (9) | 82.0 (41) | |
| White | 25.3 (168) | 2.1 (8) | 97.9 (374) | |
| HIV status | ||||
| HIV positive | 5.2 (34) | 21.5 (26) | 78.5 (95) | 41.25 (<0.001) |
| HIV negative | 94.8 (625) | 5.9 (82) | 94.1 (1318) | |
| Other STI statusb | ||||
| Tested positive for other STI(s) | 3.6 (24) | 13.6 (8) | 86.4 (51) | 3.94 (0.051) |
| Tested negative for all other STIs | 96.4 (640) | 6.8 (100) | 93.2 (1369) | |
| PrEP use | ||||
| Currently using PrEP | 4.7 (31) | 6.0 (10) | 94.0 (157) | 0.352 (0.343) |
| Not currently using PrEP | 95.3 (634) | 7.2 (99) | 92.8 (1269) | |
| Tested for syphilis in the last 6 months | ||||
| No | 32.9 (216) | 5.8 (39) | 94.2 (630) | 2.40 (0.073) |
| Yes | 67.1 (441) | 7.8 (64) | 92.2 (757) | |
| Current primary care doctor | ||||
| No | 58.3 (385) | 6.5 (51) | 93.5 (735) | 0.475 (0.278) |
| Yes | 41.7 (275) | 7.4 (53) | 92.6 (664) | |
| M (SD) | M (SD) | M (SD) | t, df (p) | |
|---|---|---|---|---|
| Condomless sexual behavior | ||||
| Oral given |
3.53 (4.45) |
3.78 (4.77) |
3.26 (4.28) |
1.2, 1533 (0.28) |
| Oral received |
3.59 (4.36) |
3.21 (3.90) |
3.38 (4.25) |
0.43, 1533 (0.67) |
| Anal insertive |
1.02 (2.82) |
1.61 (3.56) |
1.05 (2.66) |
1.60, 1533 (0.06) |
| Anal receptive |
1.08 (2.72) |
1.83 (3.55) |
1.04 (2.55) |
2.3, 1533 (0.01)
|
| Unique partners |
2.54 (2.85) |
2.57 (2.93) |
2.42 (2.84) |
−0.54, 1533 (0.60) |
| Substance use | ||||
| Club drug usec |
1.10 (3.67) |
1.58 (5.32) |
1.31 (3.84) |
−0.68, 1533 (0.25) |
| Alcohol (including to intoxication) |
3.32 (3.95) |
5.89 (6.43) |
8.0 (6.80) |
3.12, 1533 (<0.001)
|
| Marijuana | 8.21 (11.23) | 14.35 (13.07) | 8.01 (11.21) | 3.27, 1533 (<0.001) |
Numbers in RPR+ and RPR− represent total number of tests, not number of participants, so they will not reflect the sample size.
Associations significant at ≤ 0.05 are in bold.
For cell counts <5, Fisher's exact test statistic is reported.
Other STIs include oral/urethral/rectal gonorrhea, and rectal/urethral chlamydia infection.
Club drugs include cocaine, ecstasy (MDMA), methamphetamine, GHB, ketamine, and prescription stimulants.
GHB, gamma hydroxybutyrate; MDMA, methylenedioxymethamphetamine; PrEP, pre-exposure prophylaxis; RPR, rapid plasma reagin; SD, standard deviation; STI, sexually transmitted infection.
Bivariate associations with RPR reactivity
Across all three time points, 109 RPR tests (7.1%) resulted reactive. RPR titers ranged from 1:1 to 1:128. Only 50.2% of the tests that were positive for syphilis IgG were RPR reactive, suggesting that roughly half of exposures were remote or previously treated. In bivariable analysis, race/ethnicity, HIV status, other STI status, number of instances of condomless receptive anal sex, and number of days using marijuana were all significantly associated with syphilis infection when examined cumulatively (Table 1).
Associations examined by time point varied (Table 2). Syphilis RPR reactivity differed significantly by race/ethnicity at baseline and at 36 months. At all three time points, people living with HIV were significantly more likely to have syphilis. At baseline, marijuana use was significantly positively associated with syphilis RPR reactivity. At 18 months, oral sex given, condomless insertive anal sex, condomless receptive anal sex, and marijuana use were significantly positively associated with syphilis RPR reactivity. At 36 months, condomless receptive anal sex and marijuana use were significantly positively associated. At baseline and at 18 months, alcohol use was negatively associated with syphilis RPR reactivity; those with reactive syphilis test results reported significantly fewer mean days of alcohol use.
Table 2.
Sample Characteristics and Bivariate Associations with Reactive Syphilis Test Across Three Study Time Points (n = 665)
| Time 1 (baseline) |
Time 2 (18 months) |
Time 3 (36 months) |
|||||||
|---|---|---|---|---|---|---|---|---|---|
| RPR+, % (n) | RPR−, % (n) | χ2 (p)a | RPR+, % (n) | RPR−, % (n) | χ2 (p) | RPR+, % (n) | RPR−, % (n) | χ2 (p) | |
| Total | 5.0 (33) | 95.0 (631) | — | 7.8 (35) | 92.2 (412) | — | 10.2 (41) | 89.8 (361) | — |
| Race/ethnicity | |||||||||
| Hispanic/Latine | 5.2 (11) | 94.8 (201) | 13.95 (0.016) | 10.5 (14) | 89.5 (119) | 10.26 (0.068) | 10.0 (13) | 90.0 (117) | 11.20 (0.048) |
| Black | 8.2 (14) | 91.8 (156) | 9.5 (12) | 90.5 (114) | 13.7 (16) | 86.3 (101) | |||
| Asian/Pacific Islander | 4.1 (2) | 95.9 (47) | 2.8 (1) | 97.2 (35) | 3.4 (1) | 96.6 (28) | |||
| Multiracial | 4.7 (2) | 95.3 (41) | 7.1 (2) | 92.9 (26) | 13.3 (4) | 86.7 (26) | |||
| Native American/Alaska Native/other | 13.0 (3) | 87.0 (20) | 18.8 (3) | 81.2 (13) | 27.3 (3) | 72.7 (8) | |||
| White | 0.6 (1) | 99.4 (167) | 2.6 (3) | 97.4 (112) | 4.0 (4) | 96.0 (95) | |||
| HIV status | |||||||||
| HIV positive | 17.6 (6) | 82.4 (28) | 12.01 (<0.001) | 17.5 (7) | 82.5 (33) | 6.18 (0.023) | 27.7 (13) | 72.3 (34) | 18.65 (<0.001) |
| HIV negative | 4.3 (27) | 95.7 (598) | 6.6 (27) | 93.4 (382) | 7.7 (28) | 92.3 (338) | |||
| Other STI statusb | |||||||||
| Tested positive for other STI(s) | 12.5 (3) | 87.5 (21) | 3.20 (0.103) | 9.1 (2) | 90.9 (20) | 0.059 (0.522) | 23.1 (3) | 76.9 (10) | 2.58 (0.129) |
| Tested negative for all other STIs | 4.5 (29) | 95.5 (611) | 7.7 (33) | 92.3 (397) | 9.5 (38) | 90.5 (361) | |||
| PrEP use | |||||||||
| Currently using PrEP | 9.7 (3) | 90.3 (28) | 1.533 (0.195) | 3.6 (2) | 96.4 (45) | 1.537 (0.166) | 6.3 (5) | 93.8 (75) | 1.449 (0.160) |
| Not currently using PrEP | 4.7 (30) | 95.3 (604) | 8.3 (33) | 91.7 (365) | 10.7 (36) | 89.3 (300) | |||
| Tested for syphilis in the last 6 months | |||||||||
| No | 3.2 (7) | 96.8 (209) | 1.845 (0.120) | 6.3 (15) | 93.7 (231) | 0.626 (0.429) | 7.9 (17) | 92.1 (198) | 2.210 (0.094) |
| Yes | 5.7 (25) | 94.3 (416) | 8.2 (16) | 92.8 (402) | 12.4 (123) | 87.6 (163) | |||
| Current primary care doctor | |||||||||
| No | 4.9 (19) | 95.1 (366) | 0.008 (1.00) | 8.2 (17) | 91.8 (190) | 0.549 (0.288) | 7.7 (15) | 92.3 (179) | 1.701 (0.128) |
| Yes | 5.1 (14) | 94.9 (261) | 6.4 (15) | 93.6 (220) | 11.6 (24) | 88.4 (183) | |||
| M (SD) | M (SD) | t, df (p) | M (SD) | M (SD) | t, df (p) | M (SD) | M (SD) | t, df (p) | |
|---|---|---|---|---|---|---|---|---|---|
| Condomless sexual behavior | |||||||||
| Oral given |
3.30 (3.72) |
3.54 (4.50) |
0.30, 663 (0.77) |
4.49 (4.51) |
2.99 (4.01) |
−2.10, 452 (0.04)
|
3.56 (5.70) |
3.10 (4.20) |
−0.50, 414 (0.62) |
| Oral received |
2.94 (3.66) |
3.62 (4.40) |
0.88, 663 (0.38) |
3.91 (4.30) |
3.15 (3.90) |
−1.11, 452 (0.27) |
2.83 (3.75) |
3.23 (4.35) |
0.57, 414 (0.53) |
| Anal insertive |
0.89 (2.81) |
1.03 (2.82) |
0.29, 663 (0.77) |
2.49 (4.37) |
0.94 (2.13) |
2.07, 452 (0.02)
|
1.44 (3.25) |
1.23 (2.87) |
−0.44, 414 (0.66) |
| Anal receptive |
1.09 (2.97) |
1.08 (2.71) |
−0.03, 663 (0.98) |
2.23 (3.90) |
0.93 (2.13) |
1.95, 452 (0.03)
|
2.07 (3.67) |
1.11 (2.69) |
1.64, 414 (0.04)
|
| Unique partners |
2.85 (3.63) |
2.52 (2.81) |
−0.64, 663 (0.52) |
2.63 (2.83) |
2.15 (2.60) |
1.04, 452 (0.34) |
2.29 (2.38) |
2.53 (3.12) |
0.48, 414 (0.63) |
| Substance use | |||||||||
| Club drug usec |
2.55 (9.00) |
1.07 (3.21) |
−0.94, 663 (0.35) |
1.46 (2.83) |
1.42 (4.07) |
−0.07, 452 (0.94) |
0.90 (1.85) |
1.59 (4.50) |
0.98, 414 (0.33) |
| Alcohol (incl. to intoxication) |
6.42 (6.42) |
8.36 (6.77) |
1.68, 663 (0.05)
|
5.60 (6.18) |
7.76 (6.76) |
1.83, 452 (0.03)
|
5.71 (6.78) |
7.63 (6.85) |
1.72, 414 (0.09) |
| Marijuana | 12.81 (12.35) | 7.97 (11.12) | 2.43, 663 (0.02) | 16.46 (13.10) | 7.72 (11.20) | −3.83, 452 (<0.001) | 13.78 (13.68) | 11.41 (8.42) | 2.42, 414 (0.02) |
For cell counts <5, Fisher's exact test statistic is reported.
Other STIs include oral/urethral/rectal gonorrhea and rectal/urethral chlamydia infection.
Club drugs include cocaine, MDMA (ecstasy), methamphetamine, GHB, ketamine, and prescription stimulants.
PrEP, pre-exposure prophylaxis; RPR, rapid plasma reagin; SD, standard deviation; STI, sexually transmitted infection.
Binary logistic regression analysis using GEE
In binary logistic regression models, using GEE to account for repeated screening across the three time points, significant adjusted odds ratios (AOR) were observed in each model (Table 3). Each of the models controlled for race/ethnicity in addition to their primary category.
Table 3.
Binary Logistic Regression Models Using Generalized Estimating Equations (GEE) to Assess Syphilis Rapid Plasma Reagin (RPR) Reactivity Across Three Time Points (n = 665)
| AOR | 95% CI | SE | Significant | |
|---|---|---|---|---|
| Model 1: sexual behavior | ||||
| Intercept | 0.043 | 0.02–0.12 | 0.54 | <0.001 |
| Hispanic/Latine (Ref. White) | 3.70 | 1.24–10.94 | 0.56 | 0.02 |
| Black (Ref. White) | 5.12 | 1.70–15.45 | 0.56 | 0.004 |
| Asian/Pacific Islander (Ref. White) | 1.95 | 0.29–13.04 | 0.97 | 0.49 |
| Multiracial (Ref. White) | 4.05 | 0.93–17.63 | 0.75 | 0.06 |
| Native American/Alaska Native/other (Ref. White) | 9.87 | 2.46–39.59 | 0.71 | 0.001 |
| Oral sex given | 1.03 | 0.95–1.12 | 0.04 | 0.49 |
| Oral sex received | 0.88 | 0.97–0.98 | 0.06 | 0.02 |
| Condomless anal sex insertive | 1.12 | 1.01–1.25 | 0.05 | 0.03 |
| Condomless anal sex receptive | 1.09 | 1.00–1.18 | 0.04 | 0.04 |
| No. of unique sexual partners | 1.04 | 0.96–1.13 | 0.04 | 0.36 |
| Model 2: substance use | ||||
| Intercept | 0.05 | 0.02–0.14 | 0.52 | <0.001 |
| Hispanic/Latine (Ref. White) | 2.80 | 0.94–8.34 | 0.55 | 0.07 |
| Black (Ref. White) | 4.17 | 1.39–12.51 | 0.56 | 0.01 |
| Asian/Pacific Islander (Ref. White) | 1.85 | 0.31–10.91 | 0.91 | 0.50 |
| Multiracial (Ref. White) | 3.08 | 0.70–13.49 | 0.75 | 0.14 |
| Native American/Alaska Native/other (Ref. White) | 6.89 | 1.64–28.98 | 0.74 | 0.008 |
| Days using club drugs | 1.04 | 0.98–1.10 | 0.034 | 0.23 |
| Days consuming alcohol (incl. to intoxication) | 0.95 | 0.91–0.99 | 0.032 | 0.04 |
| Days using marijuana | 1.04 | 1.02–1.06 | 0.011 | <0.001 |
| Model 3: other HIV/STIs | ||||
| Intercept | 0.32 | 0.08–1.30 | 0.71 | 0.111 |
| Hispanic/Latine (Ref. White) | 3.20 | 1.09–9.41 | 0.55 | 0.035 |
| Black (Ref. White) | 4.55 | 1.54–13.43 | 0.55 | 0.006 |
| Asian/Pacific Islander (Ref. White) | 2.04 | 0.32–12.96 | 0.94 | 0.450 |
| Multiracial (Ref. White) | 3.37 | 0.77–14.73 | 0.75 | 0.107 |
| Native American/Alaska Native/other (Ref. White) | 5.14 | 1.12–23.70 | 0.78 | 0.036 |
| Positive for other STIs | 2.22 | 0.85–5.81 | 0.49 | 0.103 |
| HIV positive | 3.57 | 1.98–6.42 | 0.23 | <0.001 |
| Model 4: health care utilizationa | ||||
| Intercept | 0.187 | 0.18–1.95 | 1.20 | 0.161 |
| Hispanic/Latine (Ref. White) | 2.71 | 0.62–11.80 | 0.75 | 0.185 |
| Black (Ref. White) | 1.95 | 0.40–1.96 | 0.81 | 0.410 |
| Asian/Pacific Islander (Ref. White) | 3.20 | 0.21–49.01 | 1.39 | 0.404 |
| Multiracial (Ref. White) | 2.00 | 0.063–63.10 | 1.76 | 0.695 |
| Native American/Alaska Native/other (Ref. White) | 23.65 | 2.64–226.97 | 1.15 | 0.006 |
| Not currently using PrEP | 3.81 | 1.14–12.70 | 0.61 | 0.029 |
| Not tested for syphilis in the last 6 months | 6.68 | 2.13–20.96 | 0.58 | 0.001 |
| Does not currently have a primary care physician | 0.34 | 0.09–1.33 | 0.69 | 0.122 |
Model adjustment includes HIV status, estimate not reported.
AOR, adjusted odds ratios; CI, confidence interval; PrEP, pre-exposure prophylaxis; STI, sexually transmitted infection.
Sexual behavior
In model 1, odds of syphilis were greater among Hispanic/Latine [AOR = 3.70, confidence interval (95% CI): 1.24–10.94, p = 0.02], Black (AOR = 5.12, 95% CI: 1.70–15.14, p = 0.004), and Native American/Alaskan Native or other (AOR = 9.87, 95% CI: 2.46–39.59, p = 0.001) participants. In addition, odds of infection increased for each additional day of condomless anal insertive sex (AOR = 1.12, 95% CI: 1.01–1.25, p = 0.03) and condomless anal receptive sex (AOR = 1.09, 95% CI: 1.00–1.18, p = 0.004). Inversely, odds of syphilis RPR reactivity decreased with additional reported days of receptive oral sex (AOR = 0.88, 95% CI: 0.97–0.98, p = 0.02).
Substance use
In Model 2, odds of syphilis infection were greater among Black (AOR = 4.17, 95% CI: 1.39–12.51, p = 0.01) and Native American/Alaskan Native or other (AOR = 6.89, 95% CI: 1.64–28.98, p = 0.008) participants. Odds of syphilis infection also increased for each day of reported marijuana use (AOR = 1.04, 95% CI: 1.02–1.06, p < 0.001). Inversely, each reported day of alcohol use was associated with reduced odds of syphilis infection (AOR = 0.95, 95% CI: 0.91–0.99, p = 0.04).
HIV and other STIs
In Model 3, similar to previous models, greater odds of syphilis infection were found among participants who were Hispanic/Latine (AOR = 3.20, 95% CI: 1.09–9.41, p = 0.035), Black (AOR = 4.55, 95% CI: 1.54–13.42, p = 0.006), and Native American/Alaskan Native or other (AOR = 5.14, 95% CI: 1.12–23.70, p = 0.036). In addition, participants who previously or concurrently tested positive for HIV had greater odds of syphilis infection (AOR = 3.57, 95% CI: 1.98–6.42, p < 0.001).
Health care utilization
Lastly, in model 4, Native American/Alaskan Native or other participants had greater odds of syphilis (AOR = 23.65, 95% CI: 2.46–226.97, p = 0.006). In addition, greater odds of syphilis infection were found for participants not currently on PrEP (AOR = 3.81, 95% CI: 1.14–12.7, p = 0.029) and those who had not been tested for syphilis in the previous 6 months (AOR = 6.68, 95% CI: 2.13–20.96, p = 0.001).
Discussion
SMM are known to be at increased risk for syphilis infection, with potential drivers of this disparity including co-occurring HIV and other STIs, more frequent engagement in condomless sex, syndemic factors such as substance use, and racial inequities in health care access. We examined how these and other factors predict syphilis infection in a diverse cohort of emerging adult SMM across three time points.
Among all the syphilis tests over the 3-year study period, 7.1% were RPR reactive. Consistent with national and statewide syphilis surveillance estimates, racial/ethnic minority SMM in our cohort were more likely to have a reactive RPR result. Results from our analyses also indicated that HIV-seropositivity was positively associated with syphilis RPR reactivity at each individual time point and longitudinally, supporting extant evidence that HIV and syphilis are synergistic.4,28 We highlight this finding in light of the syphilis screening guidelines for persons living with HIV, which recommend screening at 3–6-month intervals for those who have multiple partners, engage in unprotected intercourse, or engage in illicit drug use.29
Further, syphilis risk in our cohort varied significantly based on substance use. Most notably, the number of days reported using marijuana was positively associated with syphilis RPR reactivity cross-sectionally and longitudinally, while an inverse association was seen between alcohol use and syphilis infection. However, no associations were seen when examining other illicit “club drug” use, which is well-documented as a driver of HIV seroconversion.15,16 Taken together, these findings suggest that unique typologies of substance use may exist among this cohort of SMM,14 which may help explain their differing contributions to syphilis risk.
Examining associations between sexual behavior and syphilis infection yielded similarly varied findings. While the frequency of insertive and receptive condomless anal sex was positively associated with odds of syphilis RPR reactivity, the inverse was seen for frequency of oral sex received. As with substance use, this could indicate distinct typologies of sexual behavior or suggest intentional engagement in risk reduction practices; risk of oral-genital transmission is lower for HIV, and to a lesser extent, for syphilis as well, which study participants were informed of during risk reduction counseling before HIV/STI testing. We also found no association with the number of unique sexual partners in the previous month, which is noteworthy when considering that syphilis screening guidelines frequently include the number of sexual partners among their identification criteria.29
In our model on health care engagement, those who were not currently using PrEP had greater odds of reactive syphilis RPR results. Although there is no clear or consistent evidence that PrEP use leads to increased STI risk at the individual or population levels, there remains concern about sexual risk compensation. Alternatively, researchers have pointed to this being a likely result of increased screening.21,22,27,30 When considered in light of our finding that participants who had not received recent syphilis screening were more likely to screen positive at assessment, these results support the benefits of PrEP use for overall sexual health through increased engagement in preventive care, along with the need for frequent STI screening regardless of PrEP uptake. Reducing barriers to sexual health care and increasing the availability of STI screening in settings outside PrEP and HIV care clinics should be a primary focus of public health initiatives.
Limitations
This study has several limitations to note. First, the reverse screening algorithm used to determine syphilis infection, although widely used in clinical settings, has features that affect the interpretation of our outcome in analyses. Because we used RPR testing as our research outcome, cohort participants with any syphilis exposure, including those with remote infection or who had received treatment, are undercounted. Immunoassay results would have been more sensitive in identifying these participants, but would have limited our ability to effectively model change in syphilis status over time as treponemal antibodies may remain detectable long after an infection is treated.
As a result of this, we cannot say definitively that syphilis RPR results indicate incident cases. Nor can we ascertain participants who remained serofast, such that RPR titers remained detectable after an infection had resolved, since we did not collect data on syphilis treatment. These point to the many complexities that exist with regard to staging and diagnosing syphilis both in research and clinical settings. In addition, this study relied on self-report for behavioral variables. Steps were taken to reduce the effects of social desirability and recall bias, including the use of computer-assisted self-interview and time line follow-back data collection methods.
Finally, data for these analyses were collected between July 2014 and March 2019. As such, the generalizability of our findings in a contemporary context, subject to the ongoing effects of the COVID-19 pandemic, is yet to be determined. While the increasing rate of syphilis infections among SMM antedates the emergence of COVID-19, this trend may be exacerbated due to COVID-19 limiting the availability of preventive health services, patients forgoing or delaying sexual health care, and changes in individuals' sexual behavior practices.31–33 As has been recently described, the disruptions caused by COVID-19 and the current rise in STIs among SMM presaged the emergence of the new human monkeypox virus (MPX or hMPXv) pandemic.34 As data on STI rates throughout the COVID-19 pandemic continue to emerge, adapting our screening and treatment strategies for syphilis and other STIs should be a top public health priority.
Our longitudinal analysis of syphilis infection among a diverse cohort of SMM identified several associations, some of which have not been previously investigated. These include associations with distinct sexual and substance use behaviors in emerging adult SMM, which may differ from SMM in general. This suggests a need for tailored approaches to communication regarding syphilis risk and screening, acknowledging that the number of partners alone may be an imperfect marker of risk. In addition, we found that PrEP use was associated with reduced odds of syphilis infection when controlling for screening history; an important addition to the growing evidence that standardized STI screening practices may attenuate or even invert associations between PrEP use and STI rates.
Finally, our findings support that current guidelines for syphilis screening in asymptomatic SMM, which incorporate demographic and behavioral criteria, correctly identify those at greatest risk for syphilis infection under most circumstances. However, our results also suggest that having no recent history of syphilis screening is among the strongest predictors of future infection. Universal and accessible screening for syphilis, much similar to that for HIV, may help to curtail the increasing rates of transmission.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
Research reported in this publication was supported by the National Institute on Drug Abuse (NIDA), a component of the National Institute of Health (NIH), under award number R01DA025537. P.A.D. is supported by award number T32 MH019139 (Principal Investigator, Theodorus Sandfort, PhD) from the National Institute of Mental Health. C.E.L. received support from the New Jersey Commission on Cancer Research (NJCCR), a component of the New Jersey Department of Health, as a Predoctoral Cancer Research Fellow.
References
- 1. Spicknall IH, Kreisel KM, Weinstock HS. Estimates of the prevalence and incidence of syphilis in the United States, 2018. Sex Transm Dis 2021;48(4):247–252; doi: 10.1097/olq.0000000000001364. [DOI] [PubMed] [Google Scholar]
- 2. Clement ME, Hicks CB. Syphilis on the rise: What went wrong? JAMA 2016;315(21):2281–2283; doi: 10.1001/jama.2016.7073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance 2019. 2021. Available from: https://www.cdc.gov/std/statistics/2019/default.htm [Last accessed: March 13, 2021].
- 4. Schillinger JA, Slutsker JS, Pathela P, et al. The epidemiology of syphilis in New York City: Historic trends and the current outbreak among men who have sex with men, 2016. Sex Transm Dis 2018;45(9S):S48–S54; doi: 10.1097/olq.0000000000000796. [DOI] [PubMed] [Google Scholar]
- 5. Newman DR, Matthias J, Rahman MM, et al. Repeat syphilis among HIV-infected men in Florida and Louisiana 2000–2018: Implications for screening recommendations. AIDS Patient Care STDS 2021;35(11):435–440; doi: 10.1089/apc.2021.0081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Halkitis PN, Kapadia F, Bub KL, et al. A longitudinal investigation of syndemic conditions among young gay, bisexual, and other MSM: The P18 cohort study. AIDS Behav 2014;19(6):970–980; doi: 10.1007/s10461-014-0892-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Ferlatte O, Salway T, Samji H, et al. An application of syndemic theory to identify drivers of the syphilis epidemic among gay, bisexual, and other men who have sex with men. Sex Transm Dis 2018;45(3):163–168; doi: 10.1097/OLQ.0000000000000713. [DOI] [PubMed] [Google Scholar]
- 8. Glick SN, Morris M, Foxman B, et al. A comparison of sexual behavior patterns among men who have sex with men and heterosexual men and women. J Acquir Immune Defic Syndr 2012;60(1):83–90; doi: 10.1097/QAI.0b013e318247925e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Halkitis PN, LoSchiavo C, Martino RJ, et al. Age of sexual debut among young gay-identified sexual minority men: The P18 cohort study. J Sex Res 2021;58(5):573–580; doi: 10.1080/00224499.2020.1783505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Arando M, Fernandez-Naval C, Mota-Foix M, et al. Early syphilis: Risk factors and clinical manifestations focusing on HIV-positive patients. BMC Infect Dis 2019;19(1):727; doi: 10.1186/s12879-019-4269-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Levine EC, Herbenick D, Martinez O, et al. Open relationships, nonconsensual nonmonogamy, and monogamy among U.S. adults: Findings from the 2012 National Survey of Sexual Health and Behavior. Arch Sex Behav 2018;47(5):1439–1450; doi: 10.1007/s10508-018-1178-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. MacGregor L, Kohli M, Looker KJ, et al. Chemsex and diagnoses of syphilis, gonorrhoea and chlamydia among men who have sex with men in the UK: A multivariable prediction model using causal inference methodology. Sex Transm Infect 2021;97(4):282–289; doi: 10.1136/sextrans-2020-054629. [DOI] [PubMed] [Google Scholar]
- 13. John SA, Parsons JT, Rendina HJ, et al. Club drug users had higher odds of reporting a bacterial STI compared with non-club drug users: Results from a cross-sectional analysis of gay and bisexual men on HIV pre-exposure prophylaxis. Sex Transm Infect 2019;95(8):626–628; doi: 10.1136/sextrans-2018-053591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ristuccia A, LoSchiavo C, Halkitis PN, et al. Sexualised drug use among sexual minority young adults in the United States: The P18 cohort study. Int J Drug Policy 2018;55:207–214; doi: 10.1016/j.drugpo.2018.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Guerra FM, Salway TJ, Beckett R, et al. Review of sexualized drug use associated with sexually transmitted and blood-borne infections in gay, bisexual and other men who have sex with men. Drug and Alcohol Depend 2020;216:108237; doi: 10.1016/j.drugalcdep.2020.108237. [DOI] [PubMed] [Google Scholar]
- 16. Jennings JM, Wagner J, Tilchin C, et al. Methamphetamine use, syphilis, and specific online sex partner meeting venues are associated With HIV status among Urban Black gay and bisexual men who have sex men. Sex Transm Dis 2021;48(8S):S32–S39; doi: 10.1097/OLQ.0000000000001452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Mustanski B, Morgan E, D'Aquila R, et al. Individual and network factors associated with racial disparities in HIV among young men who have sex with men: Results from the RADAR cohort study. J Acquir Immune Defic Syndr 2019;80(1):24–30; doi: 10.1097/QAI.0000000000001886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Hess KL, Hu X, Lansky A, et al. Lifetime risk of a diagnosis of HIV infection in the United States. Ann Epidemiol 2017;27(4):238–243; doi: 10.1016/j.annepidem.2017.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Force UPST. Preexposure prophylaxis for the prevention of HIV infection: US Preventive Services Task Force Recommendation Statement. JAMA 2019;321(22):2203–2213; doi: 10.1001/jama.2019.6390. [DOI] [PubMed] [Google Scholar]
- 20. Parsons JT, Rendina HJ, Lassiter JM, et al. Uptake of HIV pre-exposure prophylaxis (PrEP) in a national cohort of gay and bisexual men in the United States. J Acquir Immune Defic Syndr 2017;74(3):285–292; doi: 10.1097/QAI.0000000000001251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Montaño MA, Dombrowski JC, Dasgupta S, et al. Changes in sexual behavior and STI diagnoses among MSM initiating PrEP in a clinic setting. AIDS Behav 2019;23(2):548–555; doi: 10.1007/s10461-018-2252-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Hoornenborg E, Coyer L, Achterbergh RCA, et al. Sexual behaviour and incidence of HIV and sexually transmitted infections among men who have sex with men using daily and event-driven pre-exposure prophylaxis in AMPrEP: 2year results from a demonstration study. Lancet HIV 2019;6(7):e447–e455; doi: 10.1016/s2352-3018(19)30136-5. [DOI] [PubMed] [Google Scholar]
- 23. O'Byrne P, MacPherson P, Orser L, et al. PrEP-RN: Clinical considerations and protocols for nurse-led PrEP. J Assoc Nurses AIDS Care 2019;30(3):301–311; doi: 10.1097/JNC.0000000000000075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Saxby K, Chan C, Bavinton BR. Structural stigma and sexual health disparities among gay, bisexual, and other men who have sex with men in Australia. J Acquir Immune Defic Syndr 2022;89(3):241–250; doi: 10.1097/QAI.0000000000002851. [DOI] [PubMed] [Google Scholar]
- 25. Chandler CJ, Adams BJ, Eaton LA, et al. Intersectional experienced stigma and psychosocial syndemic conditions in a sample of Black men who have sex with men engaged in sex work (BMSM-SW) from Six US Cities. J Sex Res 2022;59(7):920–930; doi: 10.1080/00224499.2022.2072799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Henao-Martínez AF, Johnson SC. Diagnostic tests for syphilis: New tests and new algorithms. Neurol Clin Pract 2014;4(2):114–122; doi: 10.1212/01.Cpj.0000435752.17621.48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Ramchandani MS, Golden MR. Confronting rising STIs in the era of PrEP and treatment as prevention. Curr HIV/AIDS Rep 2019;16(3):244–256; doi: 10.1007/s11904-019-00446-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Blocker ME, Levine WC, St Louis ME. HIV prevalence in patients with syphilis, United States. Sex Transm Dis 2000;27(1):53–59; doi: 10.1097/00007435-200001000-00011. [DOI] [PubMed] [Google Scholar]
- 29. Tiecco G, Degli Antoni M, Storti S, et al. A 2021 update on syphilis: Taking stock from pathogenesis to vaccines. Pathogens 2021;10(11): doi: 10.3390/pathogens10111364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Morgan E, Dyar C, Newcomb ME, et al. PrEP use and sexually transmitted infections are not associated longitudinally in a cohort study of young men who have sex with men and transgender women in Chicago. AIDS Behav 2020;24(5):1334–1341; doi: 10.1007/s10461-019-02664-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Jenness SM, Le Guillou A, Chandra C, et al. Projected HIV and bacterial sexually transmitted infection incidence following COVID-19-related sexual distancing and clinical service interruption. J Infect Dis 2021;223(6):1019–1028; doi: 10.1093/infdis/jiab051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Nagendra G, Carnevale C, Neu N, et al. The potential impact and availability of sexual health services during the COVID-19 pandemic. Sex Transm Dis 2020;47(7):434–436; doi: 10.1097/OLQ.0000000000001198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Griffin M, Jaiswal J, Martino RJ, et al. Sex in the time of COVID-19: Patterns of sexual behavior among LGBTQ+ individuals in the U.S. Arch Sex Behav 2022;51(1):287–301 doi: 10.1007/s10508-022-02298-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Laurence J. The recent rise in sexually transmitted infections in the United States was a Harbinger of the new Monkeypox pandemic. AIDS Patient Care STDS 2022;36(9):333–335; doi: 10.1089/apc.2022.29009.com. [DOI] [PubMed] [Google Scholar]