Abstract
Background
Despite rising rates of syphilis among people with human immunodeficiency virus (HIV; PWH) in the United States, there is no optimal syphilis screening frequency or prioritization.
Methods
We reviewed records of all PWH in care between 1 January 2014 and 16 November 2018 from 4 sites in the Centers for AIDS Research Network of Integrated Clinical Systems Cohort (CNICS; N = 8455). We calculated rates of syphilis testing and incident syphilis and used Cox proportional hazards models modified for recurrent events to examine demographic and clinical predictors of testing and diagnosis.
Results
Participants contributed 29 568 person-years of follow-up. The rate of syphilis testing was 118 tests per 100 person-years (95% confidence interval [CI]: 117–119). The rate of incident syphilis was 4.7 cases per 100 person-years (95% CI: 4.5–5.0). Syphilis diagnosis rates were highest among younger cisgender men who have sex with men and transgender women, Hispanic individuals, people who inject drugs, and those with detectable HIV RNA, rectal infections, and hepatitis C.
Conclusions
We identified PWH who may benefit from more frequent syphilis testing and interventions for syphilis prevention.
Keywords: people living with HIV, syphilis testing, syphilis incidence
Among people with HIV in care, younger cisgender MSM and transgender women, Hispanic individuals, people who inject drugs, and those with detectable HIV RNA, rectal infections, and hepatitis C may benefit from more frequent syphilis testing and prevention interventions.
Since the early 2000s, the incidence of syphilis has been rising among gay, bisexual, and other men who have sex with men (MSM) in the United States, and more recently among women and people who inject drugs (PWID) [1–4]. Syphilis is more common in people with human immunodeficiency virus (HIV; PWH) and new cases of syphilis have increased over time among PWH engaged in care [3, 4]. Syphilis is associated with incident HIV infection, and PWH who are not virally suppressed experience a higher incidence of new and recurrent syphilis compared with those PWH who are not virally suppressed [4]. People with HIV who face barriers to regular clinic visits and antiretroviral medication adherence may also be less likely to use condoms consistently [5, 6]. In addition, some studies have found that HIV RNA levels increase during primary and secondary syphilis, which may increase the risk of onward HIV transmission [7, 8]. Syphilis is also associated with a higher incidence of sexually transmitted hepatitis C virus (HCV) infection among MSM with HIV. Syphilis may be a marker of individuals at higher risk of HCV acquisition and/or cause mucosal disruption and inflammation that facilitates HCV transmission [9].
The Centers for Disease Control and Prevention (CDC) currently recommends annual screening for syphilis among PWH and every 3–6 months among MSM at higher risk [10]. While syphilis testing among PWH in care has increased over time [3], up to one-third of sexually active MSM living with HIV have not been screened for syphilis in the prior year [11]. More frequent screening is not recommended for women with HIV, but a recent study found syphilis to be common among women living with HIV engaged in care, particularly among Black women, women who inject drugs, and women with HCV [12].
In the context of increasing syphilis diagnoses among PWH, there is an urgent need for studies to identify predictors of syphilis testing and incident cases and inform screening recommendations and delivery of behavioral and biomedical syphilis prevention [13–16]. We examined rates of syphilis testing and incident cases in a multisite clinical cohort of PWH engaged in care and sought to identify sociodemographic and clinical characteristics associated with syphilis testing and diagnosis.
METHODS
Data Source
The Centers for AIDS Research (CFAR) Network of Integrated Clinical Systems (CNICS) is a dynamic prospective observational cohort study of adult PWH in routine clinical care at 8 academic institutions across the United States [17]. Methods of data collection have been previously reported [17]. Briefly, comprehensive clinical data collected through electronic medical records and other institutional data systems undergo rigorous data-quality assessment and are harmonized in a central data repository that is updated quarterly. CNICS research has been approved by institutional review boards at each site.
We studied all PWH receiving care with at least 1 year of follow-up beginning on or after 1 January 2014 through 16 November 2018 at 4 CNICS sites with relevant data available at the time of analysis: Fenway Community Health Center/Harvard Medical School, Boston, Massachusetts; Johns Hopkins University, Baltimore, Maryland; University of Washington, Seattle, Washington; and University of California–San Diego, San Diego, California. Participant follow-up time was divided into 3-month intervals to reduce bias introduced by participants with very frequent visits (median: 13; range: 2–48) and to mirror intervals for clinic visits and syphilis testing and follow-up [10]. The observation period ended with the earliest of occurrence of death, last date of voluntary CNICS participation, or 16 November 2018.
Outcomes
Syphilis Testing
We defined syphilis testing as any non-treponemal or treponemal test performed on serum within a given 3-month follow-up interval.
Incident Syphilis
We defined a case of incident syphilis as having 1 of the 4 following criteria within a given 3-month follow-up interval:
A rapid plasma reagin (RPR) titer of 1:16 or greater at a patient’s first 3-month follow-up interval during the study period
A reactive RPR with a titer of 1:4 or greater after a nonreactive RPR in a patient with a history of a reactive RPR during the study period
A reactive RPR with a titer of 1:1 or greater after a nonreactive RPR in a patient without a history of a reactive RPR during the study period
A 4-fold or greater increase in RPR titer from 1 follow-up interval to the next follow-up interval
These criteria were chosen because they were found to have a sensitivity of 78% and a specificity of 99% for an incident syphilis case when compared with detailed medical chart review [18]. For criterion 1, we explored the impact of reducing the RPR titer cutoff to 1:8 or greater, which did not increase the sensitivity of the criteria. The addition of a reactive treponemal test to criteria 2 and 3 also did not improve the sensitivity of the criteria. Clinic-administered antibiotic treatment information (eg, intramuscular benzathine penicillin G) was not always available for incorporation into the criteria.
Covariates
Sociodemographic Characteristics
We examined age in years (18–29, 30–39, 40–49, 50–59, and 60 years and older), race/ethnicity (White, Black, Hispanic, Asian/Pacific Islander, American Indian/Alaska Native, another race or multiracial), sex-gender (cisgender man, cisgender woman, transgender man, transgender woman), CDC HIV transmission risk (heterosexual, injection drug use [IDU], MSM and MSM who use injection drugs (MSM/IDU), other/unknown), clinical site (Boston, MA; Baltimore, MD; San Diego, CA; Seattle, WA), and year of cohort entry (1995–2001, 2002–2007, 2008–2013, 2014–2018). Age was modeled as a time-varying covariate while all other sociodemographic characteristics did not vary with time. Of note, the MSM and MSM/IDU risk categories include transgender women who have sex with men.
Clinical Characteristics
We examined time-varying covariates of HIV RNA (<200 copies/mL, ≥200 copies/mL); Neisseria gonorrhoeae (GC) and/or Chlamydia trachomatis (CT) nucleic acid amplification testing (NAAT) of any anatomic site (yes, no); HCV enzyme-linked immunoassay (EIA) testing (yes, no); prior syphilis diagnosis (yes, no); positive rectal, pharyngeal, and urogenital GC NAAT (yes, no for each); positive rectal, pharyngeal, and urogenital CT NAAT (yes, no for each); and positive HCV EIA (yes, no).
Statistical Analysis
Using survival analysis methods modified for recurrent events [19], we calculated rates of syphilis testing and diagnosis with 95% confidence intervals (CIs) overall and stratified by baseline sociodemographic characteristics, including age, race/ethnicity, sex-gender, and HIV transmission risk. Because there were only 5 participants contributing 17.5 person-years of follow-up who identified as transgender men, we only estimated aggregate rates of syphilis testing and incident syphilis.
Using Cox proportional hazards regression modified for recurrent events [19] and robust standard error estimation, we calculated crude and adjusted hazard ratios (aHRs) and 95% CIs comparing rates of syphilis testing and incident syphilis by sociodemographic and clinical characteristics. We examined bivariable testing and diagnosis models including each covariate of interest. We modeled time-varying covariates differently for testing and diagnosis models. In testing models, we assessed whether GC/CT and HCV testing increased syphilis testing in the same interval (ie, concurrent sexually transmitted infection [STI]/HCV testing) and whether a detectable HIV RNA and STI/HCV diagnosis in the prior interval increased syphilis testing in the subsequent interval. In diagnosis models, we assessed whether a detectable HIV RNA and new diagnoses of HCV and site-specific GC/CT increased the risk of incident syphilis in the same interval. Finally, we assessed whether incident syphilis in the prior 3-month interval increased the risk of syphilis diagnosis in the subsequent interval.
Covariates from bivariable models with a global Wald test P < .25 were included in the multivariable models. The testing model was stratified by clinical site and total follow-up time and the diagnosis model was stratified by clinical site and total number of syphilis tests. For multivariable models, we defined statistical significance as P < .05. Log-log plots and comparisons of Kaplan-Meier observed survival curves and Cox predicted curves did not reveal violations of the proportional hazards assumption. We used STATA 16.0 (StataCorp, College Station, TX).
RESULTS
During the study period, 8455 participants contributed 29 567.5 person-years of follow-up time (median: 4 years; range: 1–5 years). Ten percent of participants were aged 16–29 years, 28.9% were non-Hispanic Black, and 18.1% were Hispanic (Table 1). Sixteen percent were cisgender women and 1.2% were transgender women. Men who have sex with men comprised 58.5% of the sample, and 58.8% entered the cohort after 2007.
Table 1.
Baseline Characteristics of People With HIV Engaged in Care: 4 US CNICS Sites, 2014–2018
| Characteristics | No. (%) |
|---|---|
| Age,y | |
| 16–29 | 859 (10.1) |
| 30–39 | 1617 (19.1) |
| 40–49 | 2482 (29.4) |
| 50–59 | 2617 (31.0) |
| 60 and older | 880 (10.4) |
| Race/ethnicity | |
| American Indian/Alaska Native | 87 (1.0) |
| Asian/Pacific Islander | 266 (3.1) |
| Black | 2444 (28.9) |
| Hispanic | 1537 (18.2) |
| White | 3954 (46.8) |
| Another race, multiracial | 167 (2.0) |
| Gender | |
| Cisgender man | 6991 (82.7) |
| Cisgender woman | 1355 (16.0) |
| Transgender man | 5 (0.06) |
| Transgender woman | 104 (1.2) |
| HIV transmission risk | |
| Heterosexual | 1783 (21.1) |
| IDU | 751 (8.9) |
| MSM | 4947 (58.5) |
| MSM/IDU | 630 (7.4) |
| Other/unknown | 344 (4.1) |
| Site | |
| Boston, MA | 1334 (15.8) |
| Baltimore, MD | 1899 (22.5) |
| San Diego, CA | 2923 (34.6) |
| Seattle, WA | 2299 (27.2) |
| Year of cohort entry | |
| 1995–2001 | 1409 (16.7) |
| 2002–2007 | 2071 (24.5) |
| 2008–2013 | 2969 (35.1) |
| 2014–2018 | 2006 (23.7) |
N = 8455.
Abbreviations: CNICS, Centers for AIDS Research Network of Integrated Clinical Systems; HIV, human immunodeficiency virus; IDU, injection drug use; MSM, men who have sex with men.
Over the entire duration of follow-up, 30 (0.3%), 1600 (18.9%), and 4803 (56.8%) participants were tested at least every 3, 6, and 12 months, respectively. Among 5577 MSM and MSM/IDU, 26 (0.5%), 1427 (25.6%), and 3789 (67.9%) participants were tested at least every 3, 6, and 12 months, respectively.
Rate of Syphilis Testing by Sociodemographic Characteristics
There were 34 989 total syphilis tests for a rate of 118 tests per 100 person-years (95% CI: 117, 119). Among cisgender men, the rate of syphilis testing was lowest among PWID aged 60 years and older and highest among MSM aged 16–29 years (Table 2). Among cisgender women, the rate of syphilis testing was lowest among American Indian/Alaska Native heterosexuals and highest among American Indians/Alaska Natives with unknown/other risk. Among transgender women, the rate of syphilis testing was lowest among those 60 years of age and older who have sex with men and who inject drugs and highest among those 30–39 years of age who have sex with men and who inject drugs.
Table 2.
Syphilis Testing and Diagnosis Rates Stratified by HIV Transmission Risk, Gender, Age, and Race/Ethnicity Among People With HIV: 4 US CNICS Sites, 2014–2018
| HIV Transmission Risk | |||||
|---|---|---|---|---|---|
| Heterosexual | IDU | MSM | MSM/IDU | Unknown/Other | |
| Rate of syphilis testing per 100 person-years (95% CI); overall rate: 118 tests per 100 person-years (95% CI: 117, 119) | |||||
| Cisgender men | |||||
| Age, y | |||||
| 16–29 | 147 (119, 181) | 146 (93, 229) | 185 (178, 193) | 164 (142, 189) | 161 (136, 192) |
| 30–39 | 104 (93, 116) | 115 (92, 144) | 161 (157, 165) | 150 (138, 161) | 135 (117, 156) |
| 40–49 | 93 (87, 100) | 95 (84, 107) | 145 (142, 149) | 124 (116, 133) | 115 (99, 134) |
| 50–59 | 85 (80, 91) | 78 (72, 84) | 125 (122, 128) | 107 (100, 115) | 101 (89, 114) |
| 60 and older | 79 (71, 87) | 62 (55, 69) | 108 (103, 112) | 103 (90, 118) | 84 (69, 102) |
| Race/ethnicity | |||||
| American Indian/Alaska Native | 104 (71, 153) | 117 (83, 166) | 142 (123, 164) | 134 (102, 177) | 125 (52, 300) |
| Asian/Pacific Islander | 79 (59, 102) | 103 (66, 159) | 143 (134, 152) | 91 (65, 128) | 105 (72, 153) |
| Black | 81 (77, 86) | 67 (62, 73) | 133 (128, 137) | 113 (102, 126) | 93 (82, 105) |
| Hispanic | 113 (105, 123) | 109 (91, 131) | 162 (158, 167) | 142 (128, 156) | 158 (140, 179) |
| White | 90 (83, 97) | 88 (80, 97) | 132 (130, 135) | 123 (118, 129) | 111 (99, 124) |
| Another race, multiracial | 97 (68, 138) | 65 (29, 144) | 157 (145, 170) | 100 (59, 169) | 78 (46, 131) |
| Cisgender women | |||||
| Age, y | |||||
| 16–29 | 96 (82, 112) | 92 (52, 162) | … | … | 99 (82, 120) |
| 30–39 | 77 (70, 84) | 75 (60, 94) | … | … | 101 (80, 129) |
| 40–49 | 70 (65, 75) | 76 (66, 87) | … | … | 77 (55, 107) |
| 50–59 | 64 (59, 68) | 61 (54, 70) | … | … | 56 (42, 74) |
| 60 and older | 54 (47, 61) | 60 (50, 72) | … | … | 63 (46, 85) |
| Race/ethnicity | |||||
| American Indian/Alaska Native | 45 (28, 71) | 66 (40, 109) | … | … | 154 (64, 370) |
| Asian/Pacific Islander | 73 (55, 98) | 89 (33, 237) | … | … | 106 (70, 161) |
| Black | 63 (59, 66) | 66 (60, 73) | … | … | 69 (58, 82) |
| Hispanic | 90 (82, 99) | 79 (54, 115) | … | … | 109 (87, 137) |
| White | 70 (63, 75) | 64 (56, 73) | … | … | 70 (55, 89) |
| Another race, multiracial | 81 (55, 117) | 102 (62, 166) | … | … | 108 (51, 226) |
| Transgender women | |||||
| Age, y | |||||
| 16–29 | … | … | 169 (131, 218) | 107 (40, 284) | … |
| 30–39 | … | … | 162 (133, 197) | 190 (123, 295) | … |
| 40–49 | … | … | 150 (125, 179) | 144 (99, 208) | … |
| 50–59 | … | … | 124 (102, 150) | 143 (96, 214) | … |
| 60 and older | … | … | 85 (55, 132) | 22 (3, 158) | … |
| Race/ethnicity | |||||
| American Indian/Alaska Native | … | … | 160 (95, 270) | no obs | … |
| Asian/Pacific Islander | … | … | 133 (84, 212) | no obs | … |
| Black | … | … | 140 (114, 172) | 118 (69, 203) | … |
| Hispanic | … | … | 137 (118, 160) | 161 (119, 219) | … |
| White | … | … | 149 (124, 181) | 122 (80, 186) | … |
| Another race, multiracial | … | … | 143 (77, 266) | no obs | … |
| Rate of incident syphilis per 100 person-years (95% CI); overall rate: 4.7 syphilis cases per 100 person-years (95% CI: 4.5, 5.0) | |||||
| Cisgender men | |||||
| Age, y | |||||
| 16–29 | 0 | 0 | 10.4 (8.8, 12.3) | 11.6 (6.7, 20.0) | 10.2 (5.1, 20.5) |
| 30–39 | 2.1 (1.0, 4.5) | 1.5 (.2, 10.5) | 11.0 (9.9, 12.2) | 10.9 (8.2, 14.4) | 8.6 (4.9, 15.2) |
| 40–49 | 1.1 (.6, 2.1) | 2.6 (1.2, 5.4) | 7.6 (6.8, 8.4) | 5.6 (4.0, 7.8) | 3.5 (1.5, 8.5) |
| 50–59 | .9 (.5, 1.7) | .9 (.4, 1.9) | 4.7 (4.2, 5.3) | 2.7 (1.8, 4.3) | 1.6 (.6, 4.3) |
| 60 and older | .2 (.02, 1.3) | 0 | 3.0 (2.3, 3.7) | 1.5 (.5, 4.6) | 2.4 (.8, 7.5) |
| Race/ethnicity | |||||
| American Indian/Alaska Native | 0 | 0 | 4.6 (2.0, 10.2) | 5.3 (1.3, 21.0) | 0 |
| Asian/Pacific Islander | 1.5 (.2, 10.4) | 5.1 (.7, 36.4) | 8.3 (6.3, 10.7) | 11.3 (4.1, 29.4) | 3.9 (.5, 27.6) |
| Black | .4 (.2, 0.9) | .2 (.05, .8) | 5.3 (4.4, 6.2) | 3.2 (1.7, 5.9) | 3.2 (1.6, 6.4) |
| Hispanic | 2.0 (1.1, 3.7) | 1.8 (.4, 7.3) | 9.7 (8.7, 10.8) | 9.0 (6.0, 13.3) | 8.2 (4.8, 14.1) |
| White | 1.3 (.6, 2.5) | 2.0 (1.1, 3.7) | 6.2 (5.7, 6.7) | 5.4 (4.3, 6.7) | 3.3 (1.7, 6.4) |
| Another race, multiracial | 3.2 (.4, 22.9) | 0 | 8.1 (5.7, 11.4) | 7.1 (1.0, 50.7) | 5.5 (.7, 39.4) |
| Cisgender women | |||||
| Age, y | |||||
| 16–29 | 0 | 0 | … | … | .9 (.1, 6.6) |
| 30–39 | .2 (.02, 1.1) | 2.0 (.5, 8.1) | … | … | 0 |
| 40–49 | .09 (.01, .7) | .4 (.05, 2.8) | … | … | 0 |
| 50–59 | .3 (.09, .8) | .2 (.3, 1.7) | … | … | 0 |
| 60 and older | 0 | .5 (.07, 3.5) | … | … | 0 |
| Race/ethnicity | |||||
| American Indian/Alaska Native | 0 | 0 | … | … | 0 |
| Asian/Pacific Islander | 0 | 0 | … | … | 0 |
| Black | .05 (.01, .3) | .4 (.09, 1.5) | … | … | 0 |
| Hispanic | .6 (.2, 1.8) | 5.8 (1.5, 23.3) | … | … | 0 |
| White | .1 (.02, 1.0) | .3 (.04, 2.0) | … | … | 1.1 (0.1, 7.5) |
| Another race, multiracial | 0 | 0 | … | … | 0 |
| Transgender women | |||||
| Age, y | |||||
| 16–29 | … | … | 8.6 (2.8, 26.6) | 0 | … |
| 30–39 | … | … | 7.9 (3.3, 19.1) | 19.0 (4.8, 76.2) | … |
| 40–49 | … | … | 7.5 (3.4, 16.8) | 0 | … |
| 50–59 | … | … | 6.0 (2.5, 14.5) | 6.0 (.8, 42.4) | … |
| 60 and older | … | … | 0 | 0 | … |
| Race/ethnicity | |||||
| American Indian/Alaska Native | … | … | 0 | no obs | … |
| Asian/Pacific Islander | … | … | 0 | no obs | … |
| Black | … | … | 4.6 (1.5, 14.2) | 9.1 (1.3, 64.5) | … |
| Hispanic | … | … | 11.8 (7.0, 19.9) | 7.7 (1.9, 30.8) | … |
| White | … | … | 2.8 (.7, 11.3) | 0 | … |
| Another race, multiracial | … | … | 0 | no obs | … |
The ellipses “ …” indicate <20 observations in the data among those with the combination of gender and HIV transmission risk.
Abbreviations: CI, confidence interval; CNICS, Centers for AIDS Research Network of Integrated Clinical Systems; HIV, human immunodeficiency virus; IDU, injection drug use; MSM, men who have sex with men; no obs, no observations.
Rate of Incident Syphilis by Sociodemographic Characteristics
There were 1406 incident syphilis cases, resulting in a rate of 4.7 cases per 100 person-years (95% CI: 4.5–5.0). Of the 1406 syphilis cases, 852 (60.6%) represented first diagnoses and 554 (39.4%) represented recurrent diagnoses during the study period. Two hundred fifty cases (17.7%), 136 cases (9.7%), 490 cases (34.8%) and 530 cases (37.7%) were based on criterion 1, 2, 3, and 4, respectively.
Among cisgender men, MSM/IDU 16–29 years of age experienced the highest rate of incident syphilis (Table 2). Hispanic women who inject drugs experienced the highest rate of incident syphilis among cisgender women. Among transgender women, those aged 30–39 years who have sex with men and inject drugs experienced the highest rate of incident syphilis followed by those who identify as Hispanic.
Sociodemographic and Clinical Predictors of Syphilis Testing
Compared with PWH aged 40–49 years, PWH aged 16–29 years had a higher rate of testing while PWH aged 50–59 years and 60 years and older had a lower testing rate (Table 3). Hispanic PWH experienced a higher testing rate than non-Hispanic White PWH and cisgender women had a lower testing rate compared with cisgender men. Compared with those with heterosexual transmission risk, MSM, MSM/IDU, and those with other/unknown transmission risk, but not PWID, had a higher rate of syphilis testing. People with HIV who entered the CNICS cohort between 2014 and 2018 had a higher rate of testing than those who entered the cohort between 1995 and 2001 (adjusted for chronological age in the multivariable model).
Table 3.
Rates of Syphilis Testing by Sociodemographic and Clinical Characteristics and Bivariable and Multivariable Syphilis Testing Models: 4 US CNICS Sites, 2014–2018
| Characteristics | Syphilis Tests | Person-Years | Syphilis Tests per 100 Person-Years (95% CI) | Crude HR (95% CI) | P | Adjusted HRa (95% CI) | P |
|---|---|---|---|---|---|---|---|
| Sociodemographic characteristics | |||||||
| Age, y | |||||||
| 16–29 | 3228 | 1924.75 | 168 (162, 173) | 1.35 (1.29, 1.42) | <.001 | 1.12 (1.07, 1.18) | <.001 |
| 30–39 | 7313 | 5140.25 | 142 (139, 146) | 1.15 (1.12, 1.20) | <.001 | 1.01 (0.98, 1.04) | .458 |
| 40–49 | 9565 | 7757.75 | 123 (121, 126) | Ref | Ref | ||
| 50–59 | 10 934 | 10 327 | 106 (104, 108) | .86 (.83, .89) | <.001 | .93 (.90, .96) | <.001 |
| 60 and older | 3949 | 4417.75 | 89 (87, 92) | .72 (.68, .76) | <.001 | .87 (.83, .91) | <.001 |
| Race/ethnicity | |||||||
| American Indian/Alaska Native | 352 | 300.5 | 117 (105, 130) | .96 (.82, 1.12) | .594 | 1.08 (.96, 1.22) | .185 |
| Asian/Pacific Islander | 1166 | 912.5 | 128 (121, 135) | 1.05 (.97, 1.14) | .256 | .96 (.89, 1.03) | .291 |
| Black | 7889 | 8593 | 92 (90, 94) | .75 (.73, .78) | <.001 | 1.01 (.98, 1.05) | .521 |
| Hispanic | 7925 | 5407.25 | 146 (143, 150) | 1.19 (1.15, 1.24) | <.001 | 1.05 (1.02, 1.08) | .003 |
| White | 16 907 | 13 820.75 | 122 (120, 124) | Ref | Ref | ||
| Another race, multiracial | 750 | 533.5 | 141 (131, 151) | 1.14 (1.05, 1.24) | .001 | .96 (.89, 1.04) | .347 |
| Gender | |||||||
| Cisgender man | 31 175 | 24 422 | 128 (126, 129) | Ref | Ref | ||
| Cisgender woman | 3282 | 4767.75 | 69 (66, 71) | .54 (.52, .57) | <.001 | .76 (.72, .80) | <.001 |
| Transgender man | 31 | 17.5 | 177 (125, 252) | 1.40 (.79, 2.46) | .249 | 1.31 (.63, 2.71) | .467 |
| Transgender woman | 501 | 360.25 | 139 (127, 152) | 1.08 (.98, 1.20) | .130 | .96 (.87, 1.06) | .424 |
| HIV transmission risk | |||||||
| Heterosexual | 4873 | 6244 | 78 (76, 80) | Ref | Ref | ||
| IDU | 1930 | 2618.25 | 74 (70, 77) | .94 (.89, 1.01) | .077 | 1.03 (.97, 1.09) | .351 |
| MSM | 24 362 | 17 433 | 140 (138, 141) | 1.76 (1.71, 1.84) | <.001 | 1.25 (1.19, 1.30) | <.001 |
| MSM/IDU | 2689 | 2167.5 | 124 (119, 129) | 1.50 (1.50, 1.69) | <.001 | 1.28 (1.20, 1.36) | <.001 |
| Other/unknown | 1135 | 1104.75 | 103 (97, 109) | 1.31 (1.20, 1.43) | <.001 | 1.08 (1.00, 1.16) | .046 |
| Year of cohort entry | |||||||
| 1995–2001 | 5011 | 5294.75 | 95 (92, 97) | Ref | Ref | ||
| 2002–2007 | 8693 | 7805.75 | 111 (109, 114) | 1.17 (1.12, 1.23) | <.001 | 1.03 (.99, 1.07) | .198 |
| 2008–2013 | 13 440 | 10 946.5 | 123 (121, 125) | 1.29 (1.23, 1.35) | <.001 | 1.02 (.98, 1.06) | .383 |
| 2014–2018 | 7845 | 5520.5 | 142 (139, 145) | 1.51 (1.44, 1.59) | <.001 | 1.05 (1.00, 1.10) | .040 |
| Time-varying HIV RNA | |||||||
| HIV RNA >200 copies/mL, prior interval | |||||||
| No | 27 102 | 24 346.5 | 111 (110, 113) | Ref | Ref | ||
| Yes | 3179 | 3107.25 | 102 (99, 106) | .95 (.92, .99) | .011 | .96 (.93, .99) | .021 |
| Time-varying STI and HCV testing | |||||||
| GC/CT NAAT (any anatomic site), current interval | |||||||
| No | 13 131 | 17 389.25 | 75 (74, 77) | Ref | Ref | ||
| Yes | 21 858 | 12 178.25 | 179 (177, 182) | 2.34 (2.28, 2.40) | <.001 | 2.00 (1.95, 2.06) | <.001 |
| HCV EIA, current interval | |||||||
| No | 24 666 | 23 278.5 | 106 (105, 107) | Ref | Ref | ||
| Yes | 10 323 | 6289 | 164 (161, 167) | 1.49 (1.46, 1.52) | <.001 | 1.02 (1.00, 1.05) | .078 |
| Time-varying STI and HCV diagnoses | |||||||
| Syphilis diagnosis, prior interval | |||||||
| No | 29 249 | 26 914.25 | 109 (107, 110) | Ref | Ref | ||
| Yes | 1032 | 539.5 | 191 (180, 203) | 1.75 (1.66, 1.84) | <.001 | 1.26 (1.20, 1.33) | <.001 |
| GC NAAT positive (any anatomic site), prior interval | |||||||
| No | 29 031 | 6806.5 | 108 (107, 110) | Ref | Ref | ||
| Yes | 1250 | 647.25 | 193 (182, 204) | 1.77 (1.69, 1.85) | <.001 | 1.13 (1.07, 1.18) | <.001 |
| CT NAAT positive (any anatomic site), prior interval | |||||||
| No | 29 005 | 26 763.75 | 108 (107, 110) | Ref | Ref | ||
| Yes | 1267 | 690 | 185 (175, 195) | 1.71 (1.63, 1.79) | <.001 | 1.08 (1.03, 1.14) | .001 |
| HCV EIA positive, prior interval | |||||||
| No | 29 966 | 27 148 | 110 (109, 112) | Ref | b | ||
| Yes | 315 | 305.75 | 103 (92, 115) | .95 (.86, 1.05) | .337 | b | |
Abbreviations: CI, confidence interval; CNICS, Centers for AIDS Research Network of Integrated Clinical Systems; CT, Chlamydia trachomatis; EIA, enzyme-linked immunoassay; GC, Neisseria gonorrhoeae; HCV, hepatitis C virus; HIV, human immunodeficiency virus; HR, hazard ratio; IDU, injection drug use; MSM, men who have sex with men; NAAT, nucleic acid amplification test; Ref, referent; STI, sexually transmitted infection.
Adjusted models are stratified by clinic site and total follow-up time contributed by each participant.
Not included in the multivariable regression model (P > .25).
The rate of syphilis testing was lower after intervals in which PWH had a detectable HIV RNA compared with after intervals in which PWH had an undetectable HIV RNA. The rate of syphilis testing was higher during intervals in which PWH tested for GC/CT compared with during intervals in which GC/CT testing did not occur. Rates of syphilis testing were higher after intervals in which PWH had a syphilis diagnosis, a positive GC NAAT, or a positive CT NAAT compared with after intervals without a syphilis diagnosis or positive GC or CT NAAT.
Sociodemographic and Clinical Predictors of Incident Syphilis
Compared with PWH aged 40–49 years, PWH aged 30–39 years had a higher rate of incident syphilis and PWH aged 50–59 years and 60 years and older had a lower rate of incident syphilis (Table 4). Hispanic PWH had a higher rate of incident syphilis than White PWH. Cisgender women had a lower rate of incident syphilis compared with cisgender men. Compared with those with heterosexual transmission risk, those with all other transmission risk had a higher rate of rate of incident syphilis; however, the aHR for PWID was not statistically significant. Compared with PWH who entered the CNICS cohort in 1995–2001, PWH who entered the cohort after 2001 experienced a higher rate of incident syphilis.
Table 4.
Rates of Incident Syphilis by Sociodemographic and Clinical Characteristics and Bivariable and Multivariable Syphilis Diagnosis Models: 4 US CNICS Sites, 2014–20188
| Characteristics | Syphilis Diagnoses | Person-Years | Syphilis Diagnoses per 100 Person-Years (95% CI) | Crude HR (95% CI) | P | Adjusted HRa (95% CI) | P |
|---|---|---|---|---|---|---|---|
| Sociodemographic characteristics | |||||||
| Age, y | |||||||
| 16–29 | 164 | 1924.75 | 8.5 (7.3, 9.9) | 1.62 (1.32, 1.99) | <.001 | 1.10 (.88, 1.39) | .389 |
| 30–39 | 439 | 5140.25 | 8.5 (7.8, 9.4) | 1.64 (1.41, 1.90) | <.001 | 1.33 (1.14, 1.55) | <.001 |
| 40–49 | 404 | 7757.75 | 5.2 (4.7, 5.7) | Ref | Ref | ||
| 50–59 | 324 | 10 327 | 3.1 (2.8, 3.5) | .60 (.51, .71) | <.001 | .81 (.69, .96) | .014 |
| 60 and older | 75 | 4417.75 | 1.7 (1.3, 2.1) | .32 (.24, .43) | <.001 | .56 (.42, .76) | <.001 |
| Race/ethnicity | |||||||
| American Indian/Alaska Native | 8 | 300.5 | 2.7 (1.3, 5.3) | .52 (.24, 1.11) | .090 | .56 (.26, 1.18) | .129 |
| Asian/Pacific Islander | 62 | 912.5 | 6.8 (5.3, 8.7) | 1.32 (.98, 1.77) | .067 | .93 (.68, 1.29) | .686 |
| Black | 172 | 8593 | 2.0 (1.7, 2.3) | .39 (.32, .47) | <.001 | 1.01 (.83, 1.22) | .912 |
| Hispanic | 416 | 5407.25 | 7.7 (7.0, 8.5) | 1.49 (1.30, 1.70) | <.001 | 1.15 (1.00, 1.33) | .052 |
| White | 713 | 13 820.75 | 5.2 (4.8, 5.5) | Ref | Ref | ||
| Another race, multiracial | 35 | 533.5 | 6.6 (4.7, 9.1) | 1.26 (.88, 1.81) | .208 | .80 (.54, 1.20) | .283 |
| Gender | |||||||
| Cisgender man | 1373 | 24 422 | 5.6 (5.3, 5.9) | Ref | Ref | ||
| Cisgender woman | 11 | 4767.75 | 0.2 (0.1, 0.4) | .04 (.02, .07) | <.001 | .27 (.13, .54) | <.001 |
| Transgender man | 0 | 17.5 | 0 | NA | NA | ||
| Transgender woman | 22 | 360.25 | 6.1 (4.0, 9.3) | 1.08 (.71, 1.79) | .736 | .86 (.56, 1.33) | .505 |
| HIV transmission risk | |||||||
| Heterosexual | 32 | 6244 | 0.5 (0.4, 0.7) | Ref | Ref | ||
| IDU | 20 | 2618.25 | 0.8 (0.5, 1.2) | 1.49 (.81, 2.76) | .202 | 2.01 (.99, 3.58) | .055 |
| MSM | 1199 | 17 433 | 6.9 (6.5, 7.3) | 13.3 (9.22, 19.3) | <.001 | 2.50 (1.67, 3.74) | <.001 |
| MSM/IDU | 122 | 2167.5 | 5.6 (4.7, 6.7) | 11.0 (7.27, 16.5) | <.001 | 2.39 (1.53, 3.74) | <.001 |
| Other/unknown | 33 | 1104.75 | 3.0 (2.1, 4.2) | 5.78 (3.32, 10.1) | <.001 | 2.00 (1.14, 3.49) | .015 |
| Year of cohort entry | |||||||
| 1995–2001 | 111 | 5294.75 | 2.1 (1.7, 2.5) | Ref | Ref | ||
| 2002–2007 | 329 | 7805.75 | 4.2 (3.8, 4.7) | 2.00 (1.57, 2.55) | <.001 | 1.29 (1.01, 1.65) | .043 |
| 2008–2013 | 591 | 10 946.5 | 5.4 (5.0, 5.8) | 2.56 (2.05, 3.21) | <.001 | 1.25 (1.00, 1.58) | .054 |
| 2014–2018 | 375 | 5520.5 | 6.8 (6.1, 7.5) | 3.22 (2.55, 4.07) | <.001 | 1.64 (1.28, 2.10) | <.001 |
| Time-varying HIV RNA | |||||||
| HIV RNA >200 copies/mL, current interval | |||||||
| No | 1181 | 25 942 | 4.5 (4.3, 4.8) | Ref | Ref | ||
| Yes | 225 | 3625.5 | 6.2 (5.4, 7.1) | 1.27 (1.09, 1.47) | .002 | 1.53 (1.29, 1.80) | <.001 |
| Time-varying STI and HCV diagnoses | |||||||
| Rectal GC NAAT positive, current interval | |||||||
| No | 1320 | 29 185.5 | 4.5 (4.3, 4.8) | Ref | Ref | ||
| Yes | 86 | 382 | 22.5 (18.2, 27.8) | 4.95 (4.01, 6.10) | <.001 | 1.66 (1.26, 2.18) | <.001 |
| Rectal CT NAAT positive, current interval | |||||||
| No | 1313 | 29 059.25 | 4.5 (4.3, 4.8) | Ref | Ref | ||
| Yes | 93 | 508.25 | 18.3 (14.9, 22.4) | 3.97 (3.20, 4.92) | <.001 | 1.47 (1.14, 1.90) | .003 |
| Pharyngeal GC NAAT positive, current interval | |||||||
| No | 1342 | 29 252.25 | 4.6 (4.3, 4.8) | Ref | Ref | ||
| Yes | 64 | 315.25 | 20.3 (15.9, 25.9) | 4.45 (3.49, 5.68) | <.001 | 1.35 (.99, 1.84) | .057 |
| Pharyngeal CT NAAT positive, current interval | |||||||
| No | 1383 | 29 452.25 | 4.7 (4.4, 4.9) | Ref | Ref | ||
| Yes | 23 | 115.25 | 20.0 (13.3, 30.0) | 4.31 (2.82, 6.57) | <.001 | 1.33 (.79, 2.26) | .285 |
| Urogenital GC NAAT positive, current interval | |||||||
| No | 1374 | 29 365.5 | 4.7 (4.4, 4.9) | Ref | Ref | ||
| Yes | 32 | 202 | 15.8 (11.2, 22.4) | 3.37 (2.42, 4.69) | <.001 | 1.15 (.80, 1.66) | .451 |
| Urogenital CT NAAT positive, current interval | |||||||
| No | 1365 | 29 329.25 | 4.6 (4.4, 4.9) | Ref | Ref | ||
| Yes | 41 | 238.25 | 17.2 (12.7, 23.4) | 3.61 (2.67, 4.89) | <.001 | 1.56 (1.11, 2.18) | .010 |
| HCV EIA positive, current interval | |||||||
| No | 1372 | 29 211.75 | 4.7 (4.4, 5.0) | Ref | Ref | ||
| Yes | 34 | 355.75 | 9.6 (6.8, 13.4) | 1.92 (1.37, 2.68) | <.001 | 1.75 (1.12, 2.74) | .014 |
| Syphilis diagnosis, prior interval | |||||||
| No | 1143 | 26 914.25 | 4.2 (4.0, 4.5) | Ref | Ref | ||
| Yes | 51 | 539.5 | 9.4 (7.2, 12.4) | 2.22 (1.68, 2.92) | <.001 | .87 (.66, 1.16) | .344 |
Abbreviations: CI, confidence interval; CNICS, Centers for AIDS Research Network of Integrated Clinical Systems; CT, Chlamydia trachomatis; EIA, enzyme-linked immunoassay; GC, Neisseria gonorrhoeae; HCV, hepatitis C virus; HIV, human immunodeficiency virus; HR, hazard ratio; IDU, injection drug use; MSM, men who have sex with men; NAAT, nucleic acid amplification test; Ref, referent; STI, sexually transmitted infection.
Adjusted models are stratified by clinic site and total number of syphilis tests contributed by each participant.
The rate of incident syphilis was higher during intervals in which PWH had a detectable HIV RNA compared with during intervals in which PWH had an undetectable HIV RNA. The rates of incident syphilis were higher during intervals in which PWH had a positive rectal GC NAAT, a positive rectal CT NAAT, and a positive urogenital CT NAAT compared with during intervals in which PWH did not experience these positive NAATs. Finally, the rate of incident syphilis was higher during an interval in which PWH had a positive HCV EIA compared with during intervals without a positive HCV EIA. In a multivariable model restricted to only MSM (excluding MSM/IDU), the rate of incident syphilis was higher during intervals in which PWH had a positive HCV EIA compared with during intervals without a positive HCV EIA (aHR: 1.84; 95% CI: 1.12, 3.03; P < .001).
DISCUSSION
Our estimates of the rate of syphilis testing and incident syphilis among PWH were higher than those previously reported among PWH engaged in care but are consistent with increasing testing and diagnosis rates over the past 2 decades [3]. Similar to prior studies, rates of syphilis testing were higher among younger PWH, cisgender men, and transgender women who have sex with men. We observed the highest rate of incident syphilis among younger MSM and transgender women who have sex with men and inject drugs, particularly among those who identify as Hispanic. These high rates of incident syphilis are likely influenced by the complex, reinforcing interactions between multiple factors, including but not limited to transphobia, homophobia, racism, and stigma related to HIV status, sexual behavior, and substance use [20–23].
Despite an association between detectable HIV RNA and an incident syphilis diagnosis, PWH with a detectable HIV RNA level at a prior visit experienced a lower syphilis testing rate than those with an undetectable HIV RNA level. Such a mismatch in screening and diagnosis may potentiate forward HIV and syphilis transmission. While syphilis may increase HIV RNA levels [7, 8], this mismatch more likely reflects barriers to syphilis screening among PWH who also face challenges in achieving durable viral suppression due to substance use, social and economic disadvantage, and discrimination based on race/ethnicity, sexual orientation, and gender identity [2, 12, 20, 22–24]. These same factors may affect condom use, sex partner selection, and conversations about STI testing with potential sex partners [5, 6]. We also observed a mismatch in screening and diagnosis rates among PWID. While PWID were not more likely to be tested for syphilis than heterosexuals, they were diagnosed with syphilis twice as frequently. This finding is consistent with an increase in syphilis cases among PWID in the United States, particularly among cisgender women [2, 12]. In our data, Hispanic cisgender women who inject drugs experienced high rates of syphilis diagnosis. This finding emphasizes the importance of integrating syphilis testing into harm-reduction and substance-use-disorder treatment programs that serve PWH.
Rates of incident syphilis were higher among PWH who were diagnosed with rectal GC or CT during the same 3-month interval. This finding has several implications. First, rectal inflammation caused by GC or CT infection may facilitate Treponema pallidum infection. Second, rectal chancres, which are often missed on clinical examination, may facilitate GC and/or CT infection of the rectal mucosa. Third, those who practice receptive anal sex may be at higher risk for syphilis compared with those who have only insertive anal sex. Fourth, those with rectal GC or CT should be routinely screened for syphilis to assess for coinfection. Finally, rectal T. pallidum NAAT may augment serologic syphilis screening among PWH, and among MSM in particular [25]. There are limitations to modeling site-specific GC/CT. Since we do not know the exact timing of T. pallidum infection in relation to GC or CT infection, we cannot draw conclusions about the directionality of this association.
Rates of incident syphilis were higher among PWH with a positive HCV EIA during the same 3-month interval. We observed this association for the entire sample and among MSM excluding MSM/IDU. Syphilis chancres may facilitate HCV transmission and/or syphilis may increase HCV RNA levels in blood and the rectal mucosa, leading to a greater probability of sexual HCV transmission [26]. Recent syphilis was associated with HCV infection among MSM living with HIV [9]. Therefore, HCV screening should be part of routine STI screening, especially among PWH with syphilis [27].
This work has important limitations. First, not all syphilis testing and diagnosis occurs within the context of HIV care. Baltimore, Maryland; San Diego, California; and Seattle, Washington, have robust local public health sexual health clinics where CNICS participants access STI testing. As syphilis testing and diagnosis outside HIV care are not necessarily captured in CNICS data, our data likely underestimate syphilis testing and diagnosis rates; matching CNICS records with public health STI surveillance data may provide more accurate rates of syphilis diagnosis and allow for evaluation of partner services, treatment, and follow-up data that may provide insight into recurrent diagnoses. Furthermore, the Baltimore, Maryland, site serves the majority (1428/2444, 58.4%) of Black PWH in our cohort; thus, this bias in syphilis testing and diagnosis rates may be particularly exaggerated among Black PWH in our sample. Possibly as a result, and in contrast to other studies, we did not find a higher rate of syphilis diagnosis among Black PWH compared with other races and ethnicities. Second, the sensitivity of the criteria used to define a syphilis diagnosis was only 78%, indicating missing syphilis diagnoses. Therefore, our estimates likely represent lower bounds of syphilis testing and diagnosis rates among PWH in this cohort. Third, the present analyses did not incorporate sexual and substance-use behaviors; future work will incorporate these data to further tailor syphilis testing recommendations. Finally, our results may not be generalizable outside well-resourced academic medical practices.
Rates of incident syphilis are high among PWH in care. Younger cisgender men and transgender women who have sex with men, PWID, Hispanic PWH, and those with detectable HIV RNA, rectal infections, and HCV are more likely to experience a syphilis diagnosis and should thus be prioritized for syphilis testing and behavioral and biomedical interventions for STI prevention [13–16].
Notes
Financial support. The Center for AIDS Research Network of Integrated Clinical Systems is funded by the National Institutes of Health, National Institute of Allergy and Infectious Diseases (grant number R24 AI067039).
Contributor Information
Timothy W Menza, Oregon Health and Science University, Portland, Oregon, USA; Oregon Health Authority, Portland, Oregon, USA.
Stephen A Berry, Johns Hopkins School of Medicine, Baltimore, Maryland, USA.
Julie Dombrowski, University of Washington School of Medicine, Seattle, Washington, USA.
Edward Cachay, University of California–San Diego School of Medicine, San Diego, California, USA.
Jodie Dionne-Odom, University of Alabama School of Medicine, Birmingham, Alabama, USA.
Katerina Christopoulos, University of California–San Francisco School of Medicine, San Francisco, California, USA.
Heidi M Crane, University of Washington School of Medicine, Seattle, Washington, USA.
Mari M Kitahata, University of Washington School of Medicine, Seattle, Washington, USA.
Kenneth H Mayer, Harvard Medical School, Boston, Massachusetts, USAand; Fenway Community Health Center, Boston, Massachusetts, USA.
References
- 1. Centers for Disease Control and Prevention. Sexually transmitted disease surveillance 2019. Atlanta, Georgia: US Department of Health and Human Services, Office of the Surgeon General, 2021. [Google Scholar]
- 2. Kidd SE. Increased methamphetamine, injection drug, and heroin use among women and heterosexual men with primary and secondary syphilis—United States, 2013–2017. MMWR Morb Mortal Wkly Rep 2019; 68:144–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Novak RM, Ghanem A, Hart R, et al. Risk factors and incidence of syphilis in human immunodeficiency virus (HIV)-infected persons: the HIV outpatient study, 1999-2015. Clin Infect Dis 2018; 67:1750–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hazra A, Menza TW, Levine K, et al. Increasing syphilis detection among patients assigned male at birth screened at a Boston community health center specializing in sexual and gender minority health, 2005-2015. Sex Transm Dis 2019; 46:375–82. [DOI] [PubMed] [Google Scholar]
- 5. Wilson PA, Kahana SY, Fernandez MI, et al. Sexual risk behavior among virologically detectable human immunodeficiency virus-infected young men who have sex with men. JAMA Pediatr 2016; 170:125–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bruce D, Bauermeister JA, Kahana SY, et al. Correlates of serodiscordant condomless anal intercourse among virologically detectable HIV-positive young men who have sex with men. AIDS Behav 2018; 22:3535–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kofoed K, Gerstoft J, Mathiesen LR, et al. Syphilis and human immunodeficiency virus (HIV)-1 coinfection: influence on CD4 T-cell count, HIV-1 viral load, and treatment response. Sex Transm Dis 2006; 33:143–8. [DOI] [PubMed] [Google Scholar]
- 8. Giacomelli A, Cozzi-Lepri A, Cingolani A, et al. Does syphilis increase the risk of HIV-RNA elevation >200 copies/mL in HIV positive patients under effective antiretroviral treatment? Data from the ICONA cohort. J Acquir Immune Defic Syndr 2021; 88(2):132–7. [DOI] [PubMed] [Google Scholar]
- 9. Medland NA, Chow EPF, Bradshaw CS, et al. Predictors and incidence of sexually transmitted hepatitis C virus infection in HIV positive men who have sex with men. BMC Infect Dis 2017; 17:185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Workowski KA. Sexually transmitted infections treatment guidelines, 2021. MMWR Recomm Rep 2021; 70. Available at: https://www.cdc.gov/mmwr/volumes/70/rr/rr7004a1.htm. Accessed 7 August 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. de Voux A, Bernstein KT, Bradley H, et al. Syphilis testing among sexually active men who have sex with men and who are receiving medical care for human immunodeficiency virus in the United States: medical monitoring project, 2013-2014. Clin Infect Dis 2019; 68:934–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Dionne-Odom J, Westfall AO, Dombrowski JC, et al. Intersecting epidemics: incident syphilis and drug use in women living with human immunodeficiency virus in the United States (2005-2016). Clin Infect Dis 2020; 71:2405–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Molina J-M, Charreau I, Chidiac C, et al. Post-exposure prophylaxis with doxycycline to prevent sexually transmitted infections in men who have sex with men: an open-label randomised substudy of the ANRS IPERGAY trial. Lancet Infect Dis 2018; 18:308–17. [DOI] [PubMed] [Google Scholar]
- 14. Grant JS, Stafylis C, Celum C, et al. Doxycycline prophylaxis for bacterial sexually transmitted infections. Clin Infect Dis 2020; 70:1247–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Mustanski B, Garofalo R, Monahan C, 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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Madkins K, Moskowitz DA, Moran K, et al. Measuring acceptability and engagement of the keep it up! Internet-based HIV prevention randomized controlled trial for young men who have sex with men. AIDS Educ Prev 2019; 31:287–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Kitahata MM, Rodriguez B, Haubrich R, et al. Cohort profile: the Centers for AIDS Research Network of Integrated Clinical Systems. Int J Epidemiol 2008; 37:948–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Menza TW, Levine K, Grasso C, et al. Evaluation of 4 algorithms to identify incident syphilis among HIV-positive men who have sex with men engaged in primary care. Sex Transm Dis 2019; 46:e38–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Amorim LDAF, Cai J.. Modelling recurrent events: a tutorial for analysis in epidemiology. Int J Epidemiol 2015; 44:324–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Mayer KH. Old pathogen, new challenges: a narrative review of the multilevel drivers of syphilis increasing in American men who have sex with men. Sex Transm Dis 2018; 45(9S Suppl 1):S38–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Allan-Blitz L-T, Menza TW, Cummings V, et al. Multifactorial correlates of incident bacterial sexually transmitted infections among black men who have sex with men recruited in 6 US cities (HPTN 061). Sex Transm Dis 2021. Available at: https://journals.lww.com/stdjournal/Abstract/9000/Multifactorial_Correlates_of_Incident_Bacterial.97717.aspx. Accessed 3 May 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. 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; 45:163–8. [DOI] [PubMed] [Google Scholar]
- 23. Semaan S, Sternberg M, Zaidi A, et al. Social capital and rates of gonorrhea and syphilis in the United States: spatial regression analyses of state-level associations. Soc Sci Med 1982 2007; 64:2324–41. [DOI] [PubMed] [Google Scholar]
- 24. Menza TW, Hixson LK, Lipira L, et al. Social determinants of health and care outcomes among people with HIV in the United States. Open Forum Infect Dis 2021; 8: ab330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Golden M, O’Donnell M, Lukehart S, et al. Treponema pallidum nucleic acid amplification testing to augment syphilis screening among men who have sex with men. J Clin Microbiol 2019; 57:e00572–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Foster AL, Gaisa MM, Hijdra RM, et al. Shedding of hepatitis C virus into the rectum of HIV-infected men who have sex with men. Clin Infect Dis 2017; 64:284–8. [DOI] [PubMed] [Google Scholar]
- 27. Hill LA, Martin NK, Torriani FJ, et al. Screening for sexually transmitted infections during hepatitis C treatment to predict reinfection among people with HIV. Open Forum Infect Dis 2021; 8:ofaa643. [DOI] [PMC free article] [PubMed] [Google Scholar]