Sexually Transmitted Infections Detected During and After Incarceration Among People with HIV: Prevalence and Implications for Screening and Prevention

Abstract

Background

Incarceration and HIV are associated with sexually transmitted infections (STIs), however little is known about STI prevalence among people living with HIV (PLWH) during and after incarceration.

Methods

Electronic medical records from the Dallas County Jail (DCJ) and community HIV clinics were reviewed to determine the frequency and results of testing for gonorrhea, chlamydia, syphilis, and Hepatitis B (HBV) among PLWH incarcerated in DCJ between 2010–2013. HIV viral loads (VL) and evidence of STI symptoms and treatment were also collected.

Results

During 2473 incarcerations, 6/190 (3%) tests were positive for gonorrhea, 7/190 (4%) for chlamydia, 231/1082 (21%) for syphilis, of which 53 (23%) were new diagnoses, and 48/1005 (5%) for HBV surface antigen (HBsAg). Among 1631 releases to the community, 808 followed-up in community clinics, where 21/553 (4%) tests were positive for gonorrhea, 23/555 (4%) for chlamydia, 150/808 (19%) for syphilis, of which 31 (21%) were new diagnoses, and 24/421 (6%) for HBsAg. The majority of new STI cases, 51/64 (80%) in jail and 43/56 (77%) in the community, had a concurrent detectable (>200 copies/mL) HIV VL.

Conclusions

Testing for gonorrhea and chlamydia was low, particularly in jail, which was attributed to testing protocols. High proportions of PLWH tested positive for syphilis and HBV infection in both settings. The majority of patients with active STIs had a detectable HIV VL. Routine, opt-out screening for STIs for PLWH during and after incarceration has the potential to identify a high proportion of STIs and improve secondary HIV prevention.

Summary

An STI study in people living with HIV during and after incarceration found low frequency of testing for GC/CT, high syphilis positivity, and concurrent detectable HIV viral load with acute STIs.

Introduction

The prevalence of Human Immunodeficiency Virus (HIV) infection is three times higher in jails and prisons than in the general population (1–2), and rates of sexually transmitted infections (STIs), such as gonorrhea (GC), chlamydia (CT), syphilis, and hepatitis B (HBV) are also high among incarcerated individuals (3–7). Incarceration presents a chance to screen individuals for both HIV and STIs, which are often asymptomatic and therefore go undetected and untreated, increasing the risk of transmission to others. In addition, incarcerated individuals may have limited access to routine healthcare in the community.

While multiple studies have analyzed STI prevalence in the incarcerated population and separately in people living with HIV (PLWH) (1–7), there are limited data on STIs among incarcerated or recently released PLWH. An acute STI is an indication of recent condomless sexual activity, and in PLWH with a detectable HIV viral load, increases the risk of HIV transmission to others. Incarceration can therefore be a critical time for STI screening and counseling on safer sex practices and adherence to HIV medications. In addition, few studies include clinical data from both the criminal justice system and community healthcare settings for incarcerated individuals, missing an opportunity to measure linkage to care and HIV/STI outcomes in the post-incarceration period. Furthermore, many STI prevalence studies are based solely on laboratory testing data and lack clinical information about symptoms, thereby missing important information that can inform screening strategies.

We sought to determine STI positivity for GC, CT, syphilis, and HBV in PLWH during jail incarceration and separately in recently released PLWH accessing HIV care in the community. We further characterized whether individuals with an STI were symptomatic, if STI treatment was provided, and for syphilis, whether this represented a new infection. In addition, we explored whether or not an individual had a detectable HIV viral load at the time of a new STI diagnosis. We anticipate that these findings will inform future STI screening strategies during incarceration and after release for people with HIV.

Materials and Methods

Participants and Procedures

This retrospective study involved a detailed chart review of all incarcerations of individuals known to have HIV who were detained at the Dallas County Jail (DCJ) between June 1, 2010 and November 30, 2013. In addition, we examined data from Dallas community HIV clinics for all initial clinic visits after jail release during the study time period. In the DCJ, individuals were identified as PLWH if disclosed by the individual during medical intake, known from a prior incarceration, or if the individual tested positive for HIV while in jail. At the time of this study, HIV testing in the jail was offered by request or as opt-in testing on days where testing agencies were present. Individuals were excluded if they were under the age of 18. This study was approved by the Institutional Review Board at UT Southwestern Medical Center.

DCJ is the 7^th largest jail in the US, with an average daily census of 6301, 255 admissions per day, and more than 100,000 admissions per year. The average length of stay is 25 days, although 46% are released within 72 hours (C. Berry, personal communication, Dec. 4, 2018). Standard of care at DCJ is that all individuals with acute or chronic health needs are evaluated within 24 hours of jail entry by a general medical provider. For PLWH, a clinic visit is scheduled with a jail HIV provider within one week, during which routine laboratory tests are sent, including syphilis (rapid plasma reagent [RPR]) titers, HIV viral load (VL), an acute hepatitis panel (including HBV surface antigen [HBsAg]), a baseline chemistry panel, complete blood counts, and CD4 count. GC/CT urine tests are sent by request (patient or provider may request), but not performed routinely. Standard of care at the community HIV clinics includes syphilis testing (RPR titers), HIV VL, HBV serology (surface antigen and antibody), CD4 count, and GC/CT urine testing for all individuals completing a new intake or individuals who have not been seen in over a year. For individuals recently released from jail with established care at the clinic, GC/CT testing is performed on a case-by-case basis, assessed by the provider.

Data Collection

Data were collected from the DCJ electronic medical record (EMR) (Pearl, Business Computer Applications, Atlanta, GA), Parkland Health and Hospital Systems EMR (Epic, Epic Systems Corporations, Verona, WI), and Prism Health North Texas EMR (Centricity, General Electric Company, Boston, MA). These clinics were chosen because they provide HIV care to the majority of uninsured patients in Dallas, and 90% of patients leaving jail receive care in this system. Records were matched with Ryan White Services Report (RSR) data from both Parkland and Prism Health. Matching of data was performed using an encrypted unique client identifier (eUCI), an identifier created from the participant’s name, date of birth, and gender using a hashing trap door algorithm to create a unique 40-digit string of letters and numbers (8). Data were validated though EMR review to confirm HIV status and inclusion criteria. Incarceration dates were obtained from the Adult Information System from the DCJ, and if an individual had multiple incarcerations during the study period, data from each incarceration were included.

Independent variables

Baseline characteristics were obtained from these data sources and included race/ethnicity, gender, age, HIV risk factor, socioeconomic factors, behavioral health factors, and healthcare utilization before and after incarceration, described in detail elsewhere (9). Age was calculated from the study end date (November 30, 2013) using date of birth from the DCJ EMR. Race/ethnicity was coded as non-Hispanic black, non-Hispanic white, and Hispanic. Gender was defined as male, female, or transgender, though due to the small sample size of the transgender subgroup (n=14, all male to female), sex at birth was used for this data analysis. HIV risk factor was coded as history of injection drug use (IDU), men who have sex with men (MSM), and heterosexual/other. Individuals with multiple risk factors were coded by highest transmission risk (IDU, then MSM, then heterosexual/other). Physical/sexual abuse was self-reported and includes any lifetime history of being abused. Substance use was self-reported and obtained from jail EMR notes as a binary composite variable of stimulant/opioid use for reported cocaine, crack cocaine, heroin, opiate, or methamphetamine use in the 12 months prior to incarceration. Methamphetamine use was also evaluated separately. Serious mental illness was defined as psychosis, bipolar disorder, and/or schizophrenia and was coded as yes/no. Self-reported information on HIV care in the 12 months prior to incarceration, including care at an HIV clinic, prior prescription of ART, and prior adherence to ART, was obtained from records.

Primary outcome variables

Detailed review of the DCJ EMR was conducted to tabulate frequencies of testing for GC/CT (urine sample, nucleic acid amplification test [NAAT]), syphilis (RPR), HBV (HBsAg), and HIV VL. A positive NAAT for GC/CT was considered an active/new infection if it was the first positive in the system. Syphilis was considered a new infection if: (a) it was the first positive in our system, (b) it was 4-fold higher than a previous titer, or (c) the positive titer followed a previously negative titer. If there was a positive titer but the above criteria were not met, it was not considered a new infection. HIV VL conducted within one month of diagnosis of a new STI was included in the analyses and defined as detectable if ≥ 200 copies/mL. If the patient reported urogenital symptoms (dysuria, discharge, lesions) at time of STI diagnosis, it was recorded as a symptomatic infection; all others were considered asymptomatic. An STI was recorded as treated if the patient received treatment according to CDC guidelines (ceftriaxone and azithromycin for GC, azithromycin for CT, benzathine penicillin G for syphilis) within one month of diagnosis (10). For all releases where an individual followed-up in a community clinic after release, results of STI (GC/CT, syphilis, HBsAg) and HIV VL testing were similarly obtained from the community clinics’ EMR.

Data Analysis

Baseline covariates were summarized using median for age and frequencies/proportions for all other categories. Proportions of tests completed and test results from DCJ and community HIV clinics were presented for each STI as absolute values and proportions. Our main unit of analysis was incarceration rather than individuals, and incarcerations were treated as unique events. Because HIV labs, including RPR, are ordered at the time of a patient visit, RPR testing was used as a proxy for a jail HIV provider visit. All analyses were completed using SAS, version 9.4 (SAS Institute, Cary, NC).

Results

There were 2473 incarcerations involving 1696 unique PLWH that met study criteria and were included in the final analysis. Baseline characteristics of the study group are described in Table 1a. Incarcerations during which an individual tested positive for multiple STIs were listed as positive in each corresponding category. Length of incarceration ranged from 0 to 597 days, with a median of 9 and mean of 34.

Table 1a:

Baseline Characteristics of Incarcerations of PLWH with GC, CT, or Syphilis from Dallas County Jail Records

	Unique Individuals n=1696		Total Incarcerations n=2473		Incarc. with Provider Visit n=1082		GC or CT Dx n=11		New Syphilis Dx n=53		Positive RPR n= 231
Age, Median (SD)	39.1 (10.6)		38.5 (10.5)		41.3 (10.4)		35.5 (10.1)		35.7 (8.3)		39.6 (10.4)
	n	%	n	%	n	%	n	%	n	%	n	%
Male Gender	1357	80.0	1919	77.6	832	76.9	5	45.5	50	94.3	184	79.7
Race/Ethnicity
Black (Non-Hispanic)	1072	63.2	1614	65.3	693	64.0	11	100	31	58.5	152	65.8
White (Non-Hispanic)	363	21.4	515	20.8	241	22.3	0	0	13	24.5	41	17.7
Hispanic	258	15.2	341	13.8	147	13.6	0	0	9	17.0	38	16.5
HIV Risk Factor
IDU	183	10.8	295	11.9	169	15.6	0	0	6	11.3	34	14.7
MSM (non IDU)	551	32.5	804	32.5	410	37.9	2	18.2	40	75.5	134	58.0
Heterosexual	962	56.7	1374	55.6	503	46.5	9	81.8	7	13.2	63	27.3
Ever Married	421	24.8	617	24.9	323	29.9	2	18.2	13	24.5	52	22.5
Physical/Sexual Abuse	393	23.2	632	25.6	322	29.8	5	45.5	12	22.6	75	32.5
Stable/Permanent Housing	429	25.3	594	24.0	212	19.6	5	45.5	14	26.4	54	23.4
Employed	172	10.1	236	9.5	114	10.5	1	9.1	8	15.1	27	11.7
Behavioral Health
Stimulant/Opioid Use	722	42.6	1192	48.2	650	60.1	8	72.7	36	67.9	152	65.8
Meth Use	249	14.7	389	15.7	217	20.1	1	9.1	19	35.8	59	25.5
Serious Mental Illness	514	30.3	873	35.3	447	41.3	4	36.4	8	15.1	73	31.6
Prior HIV Care
Before Incarceration	1061	62.6	1572	63.6	810	74.9	7	63.6	36	67.9	167	72.3
Adherence to ART	752	44.3	1040	42.1	533	49.3	3	27.3	25	47.2	112	48.5
Prescribed ART	1115	65.7	1619	65.5	831	76.8	7	63.6	38	71.7	167	72.3

Of 2473 incarcerations, 1082 (44%) had a provider visit and testing was completed for GC/CT in 190 (18%), with 6 (3%) tests positive for GC and 7 (4%) for CT (Table 2a). Of those tested, 75 (40%) were male. Of positive tests, 100% of individuals were black, 46% were male, and 82% were heterosexual. Of 1082 incarcerations with syphilis testing, 231 (21%) tests were positive by RPR, and 53 (23%) of these represented new infections. Individuals with new syphilis infections were predominately male (94%), black (59%), MSM (76%), and more frequently reported methamphetamine use (36% compared to 16% in the entire study group). There were 1005 (41%) incarcerations with tests for HBsAg, with 48 (5%) positive tests, representing 42 unique individuals. Of these, the majority were male (91%), black (67%), MSM (55%), and reported opioid/stimulant use (64%).

Table 2a:

STI Prevalence/Testing for GC, CT, and Syphilis of PLWH with an HIV Provider Visit During Incarceration from Dallas County Jail Records n = 1082 incarcerations

	# Tested	# Positive	% Positive	% Tested	% Symptoms
Gonorrhea	190	6	3.2	17.6	66.7
Chlamydia	190	7	3.7	17.6	57.1
Syphilis (+ RPR)	1082	231	21.3	100	--
Syphilis (New)	1082	53	4.9	100	1.9
Hepatitis B sAg	1005	48	4.8	92.9	--

Of these incarcerations, 669 were released to prison and 173 planned care outside publicly funded Dallas community HIV clinics and not evaluated further (Figure 1). This left 1631 releases, representing 1291 unique individuals, eligible for post-incarceration evaluation in community HIV clinics. Demographics of this study group are found in Table 1b. Of these releases, 808 (49%) linked to care, a majority (93%) within a year after release, with a range of 0 to 837 days, median of 63 and mean of 113. In 553 (69%), testing was performed for GC, of which 21 (4%) were positive; of 555 (69%) tests for CT, 23 (4%) were positive (Table 2b). Those with positive tests for GC/CT were majority male (58%), black (83%), and heterosexual (78%). An RPR was obtained at all 808 clinic visits following release, with 150 (19%) positives, and 31 (21%) of these represented new infections. Of new syphilis infections, a majority of individuals were male (87%), black (61%), and MSM (55%). Of 421 (26%) clinic visits with HBsAg testing, 24 (6%) tests, representing 24 unique individuals, were positive. Nine of these individuals had a positive HBsAg in jail and the community, and fifteen were not tested in jail. One individual tested positive in jail and negative in the community, and no individuals tested negative in jail and positive in the community. Individuals with a positive HBsAg test were predominately male (96%), black (63%), heterosexual (54%), and 46% reported stimulant/opioid use.

Figure 1:

Flowchart of Total Incarcerations and Release Dispositions among People Living with HIV Incarcerated at the Dallas County Jail

Table 1b:

Baseline Characteristics of PLWH Released to the Community with GC, CT, or Syphilis from Community HIV Clinic Records

	Unique PLWH Released n=1219		Total Incarcerations Released n=1631		Releases with Follow-Up n = 808		GC or CT Dx n=36		New Syphilis Dx n=31		Positive RPR n= 150
Age, Median (SD)	38.5 (10.6)		38.2 (10.6)		39.4 (10.7)		31.6 (8.9)		35.9 (10.0)		38.2 (10.6)
	n	%	n	%	n	%	n	%	n	%	n	%
Male Gender	952	78.1	1244	76.3	606	75.0	21	58.3	27	87.1	125	83.3
Race/Ethnicity
Black (Non-Hispanic)	792	65.0	1103	67.6	558	67.8	30	83.3	19	61.3	100	66.7
White (Non-Hispanic)	247	20.3	306	18.8	169	20.5	2	5.6	5	16.1	21	14.0
Hispanic	177	14.5	219	13.4	96	11.7	4	11.1	7	22.6	29	19.3
HIV Risk Factor
IDU	126	10.3	178	10.9	87	10.6	2	5.6	1	3.2	17	11.3
MSM (non IDU)	355	29.1	485	29.7	262	31.8	6	16.7	17	54.8	73	48.7
Heterosexual	738	60.5	968	59.4	474	57.6	28	77.8	13	41.9	60	40.0
Ever Married	284	23.3	384	23.5	173	21.0	10	27.8	2	6.5	25	16.7
Physical/Sexual Abuse	282	23.1	400	24.5	198	24.1	11	30.6	12	38.7	43	28.7
Stable/Permanent Housing	328	26.9	418	25.6	192	23.3	6	16.7	8	25.8	43	28.7
Employed	102	8.4	133	8.2	69	8.4	3	8.3	4	12.9	19	12.7
Behavioral Health
Stimulant/Opioid Use	499	40.9	730	44.8	361	43.9	13	36.1	13	41.9	70	46.7
Meth Use	165	13.5	226	13.9	128	15.6	3	8.3	5	16.1	22	14.7
Serious Mental Illness	375	30.8	567	34.8	309	37.5	10	27.8	7	22.6	38	25.3
Prior HIV Care
Before Incarceration	720	59.1	969	59.4	424	51.5	19	52.8	20	64.5	108	72.0
Adherence to ART	511	41.9	645	39.5	275	33.4	9	25.0	11	35.5	76	50.7
Prescribed ART	767	62.9	1012	62.0	465	56.5	16	44.4	16	51.6	99	66.0

Abbreviations: PLWH= People living with HIV; GC= Gonococcal infection; CT= Chlamydia trachomatis; Dx= diagnosis; RPR= rapid plasma regain; MSM= men who have sex with men; IDU= injection drug use; Meth= methamphetamine; ART= antiretroviral therapy.

Table 2b:

STI Prevalence/Testing for GC, CT, and Syphilis for PLWH with Post-Incarceration Follow-Up from Community HIV Clinic Records n = 808 releases

	# Tested	# Positive	% Positive	% Tested	% Symptoms
Gonorrhea	553	21	3.8	68.4	57.1
Chlamydia	555	23	4.1	68.7	13.0
Syphilis (+ RPR)	808	150	18.6	100	--
Syphilis (New)	808	31	3.8	100	32.3
Hepatitis B sAg	421	24	5.7	52.1	--

Abbreviations: STI= sexually transmitted infections; GC= gonococcal infection; CT= Chlamydia trachomatis; PLWH= People Living with HIV; RPR=rapid plasma regain; sAg= surface antigen.

There were 23 incarcerations in the jail where an individual identified as transgender, all male to female, representing 14 unique individuals. Testing was positive for a new STI in 4/9. Amongst 14 transgender individuals released to the community who presented for HIV care, 8 were tested for STIs, of whom one tested positive.

Overall, the proportion of individuals who received STI treatment was high, with 86% for CT, 100% for GC, and 97% for syphilis in DCJ, and 100% for GC, 100% for CT, and 94% for syphilis in community clinics.

In 64 incarcerations and 64 community releases, a new STI (GC/CT/or new syphilis) was detected. The majority had HIV VL drawn within one month of diagnosis which was detectable (≥200 copies/mL) in a majority of new STI diagnoses in both settings (Figure 2).

Figure 2:

51/64 (79.7%) of new STI diagnoses in the jail and 43/56 (76.8%) in the community occurred in the setting of a detectable (≥ 200 copies/mL) HIV viral load within one month of diagnosis, while 13/64 (20.3%) of STI diagnoses in jail and 13/56 (23.2%) in the community occurred in the setting of an undetectable (<200 copies/mL) HIV viral load. In the community, 8/64 (12.5%) of new STI diagnoses were made with no data on HIV viral load. All viral loads were drawn in the same setting that the STI was diagnosed in.

Three individuals tested positive for a new STI in both settings. Of 36 releases with a positive GC/CT test in the community, 35 were not tested in jail and 1 tested negative. Of 31 releases with a new diagnosis of syphilis in community HIV clinics, 21 were not tested in jail, 8 tested negative in jail, and 2 tested positive in both settings.

Discussion

We identified that a low proportion of PLWH were tested for GC/CT, particularly in jail, where this testing was offered by request or based on symptoms. Syphilis testing was more widespread and showed high positivity, with a positive RPR detected among nearly 1 in 5 incarcerations, of which 20% represented a recent infection. New syphilis infections detected in jail and in the community were more common among men, black individuals, and MSM. We also found that a majority who tested positive for an acute STI (GC, CT, or a new diagnosis of syphilis) during incarceration and in community follow-up had a detectable HIV VL at time of diagnosis. We propose changes in screening strategies that can improve STI detection in justice-involved PLWH.

Testing for GC and CT in both settings was low, highlighting a missed chance for intervention in a high-risk population. National GC and CT rates have been rising and are persistently high among PLWH and incarcerated individuals (11–12). GC/CT positivity from triple site (urine/urethral, oropharynx, rectum) nucleic acid amplification testing in PLWH was estimated to be 10.−15.5% for CT and 10.0–20.9% for GC (13–14). Positivity was lower in our study group, however testing overall was low, and at the time of this study, both settings offered only urine testing. Limiting testing to the urinary tract misses more than half of GC/CT infections in MSM living with HIV, who often have extragenital (rectal/pharyngeal) infections (13–15). The proportion of PLWH tested for GC/CT was considerably lower than those tested for syphilis in both settings, which we attributed to differences between testing methods. RPR testing for syphilis is a blood test and can be added on to existing blood draws, whereas GC/CT testing requires collecting a urine sample, and optimally a rectal and pharyngeal sample. In the jail setting, this requires additional security time as an officer must escort the participant to the bathroom. In our study group, testing for GC/CT in community HIV clinics was markedly higher than in DCJ, which we attributed to routine order sets that include GC/CT screening of all new patients or those returning after >12 months, whereas in the jail, screening for GC/CT was symptom-guided or by request. Studies have shown transitioning from opt-in to opt-out testing in a jail increased the number of GC/CT diagnoses by 4 to 16 fold (16–17). Most individuals who tested positive for a new STI in the community were not tested in jail, which may indicate a missed chance to receive treatment in jail or an infection acquired post-release. Additionally, only 44% of incarcerations were long enough to include an HIV provider visit. Including routine, opt-out, triple-site STI testing earlier in incarceration, such as in the medical intake visit within 24 hours of entry, would vastly increase testing and treatment opportunities and therefore public health impact.

New diagnoses of syphilis made up a majority of all new STIs in this study, and prevalence estimates in our study group were higher than other studies in PLWH. The rate of primary and secondary syphilis infections has been estimated to be 0.013% (13 per 100,000) in the Dallas general population and 1.8% in PLWH nationally (11,13,18), and early syphilis infections were estimated to be 1.6% in the overall MSM population incarcerated at the Los Angeles county jail between 2000–2005 (19). In our study, the proportion of PLWH diagnosied with a new syphilis infection was 4.9% in the jail (9.8% among MSM tested during incarceration) and 3.8% in the community after release (6.5% among MSM). Most new infections were asymptomatic, and a majority of individuals had a detectable HIV VL at time of diagnosis, which is particularly salient since ulcerative STIs such as syphilis increase HIV transmission risk (20–21). These high numbers of syphilis infections occur in the setting of concerning local, as well as national, trends of increasing syphilis incidence since 2000 (11). A disproportionate number of new cases are found in MSM, and nationally, almost half of MSM with syphilis are co-infected with HIV (12). We identified a high proportion of persons with syphilis infection, particularly in MSM, underscoring the importance of universal screening for syphilis in PLWH who are incarcerated and recently released, as well as providing counseling on safe sexual practices to prevent HIV transmission.

HBV is another STI that disproportionately affects PLWH, with prevalence of HBV exposure estimated to be as high as 50–90% (22) and proportion with active infection (by positive HBsAg or detectable HBV DNA) estimated at 8.4% in one large outpatient study (23). In our study group, the proportion with active HBV, measured by HBsAg, was 4.8% in the jail and 5.7% in the community. Duration of infection with HBV and transmission mean was not determined as part of this study. Regardless of the origin of infection, HBV status is clinically relevant when choosing antiretroviral therapy (ART), as regimens which treat both HIV and HBV are required for co-infected patients. This is particularly important in the era of rapid ART, where individuals may be started on ART prior to availability of baseline laboratory results. Furthermore, HBV is a vaccine-preventable disease, and DHHS guidelines recommend HBV vaccine series for all people with HIV (24). Incarceration provides an additional chance for HBV screening and to vaccinate those who are not immune to HBV.

In addition to evaluating the implementation and results of STI testing, we characterized risk factors for STIs in currently and recently incarcerated PLWH. A majority of STIs were diagnosed in black individuals, although our study group was predominately black (65%). Nationally, GC, CT, and syphilis are all found disproportionately in black individuals (11), and studies investigating PLWH or justice-involved individuals also support this finding (25–26). While a majority of our study group was male, females were disproportionately represented in GC/CT testing and cases, which is partly attributed to routine testing during well-woman exams. Additionally, single-site urine testing may have missed extragenital infections in MSM. Our syphilis cases were mostly found in MSM and in the jail were found twice as often in those who reported methamphetamine use, an association attributed to increased risky sexual behavior in methamphetamine users (27). These demographic trends and risk factors associated with STIs reinforce the importance of comprehensive and routine STI screening in the incarcerated and recently released individuals.

An elevated HIV VL was detected in 77–80% of new STIs diagnosed in jail and after release, which is considerably higher than what is reported in the literature—(35–44%) (18, 26). The high percentage of detectable HIV VL in our study group highlights the additional increased risk of HIV transmission, particularly in justice-involved individuals. In our study group, 44% reported prior adherence to ART, however treatment interruption is common after incarceration. Incarceration can be an opportune time for safer sex counseling as well as for re-engagement in HIV care and ART adherence counseling. Additionally, screening and treatment of STIs for incarcerated persons may reduce STIs in the general community (28).

Our study has several limitations. First, this was a retrospective study, and analyses were limited to pre-existing data that were obtained from chart review. However, data were collected from multiple sources, including jail release data, jail medical records, community records from several sites, case manager records, and RSR. Second, the proportion of people undergoing GC/CT testing was low, and there was no extragenital testing available, limiting conclusions about the true prevalence of these infections in this population. Third, due to short incarcerations and low numbers achieving timely post-release linkage to HIV care, not all PLWH with current or recent incarceration had a medical visit during the study time period or a chance for STI testing, leading to missing data. Fourth, we did not have access to health department data on STIs or HIV and therefore are limited by the data that we had access to. However, per routine clinical practice, in the setting of a positive RPR and unknown prior syphilis testing/treatment, the HIV provider from the jail will contact the health department to determine prior syphilis testing and treatment, and this is recorded in the EMR. Though we did not directly utilize results from the health department as primary data, any results recorded in the EMR were utilized to determine if infections were new. Fifth, our main unit of analysis was incarceration rather than individuals. However, characteristics of individuals closely matched incarcerations. Prior sensitivity analyses showed similar results when evaluated by incarceration and when evaluated by individual (9). Lastly, our study involved a single jail site and multiple community clinics, all in the Dallas area, which may limit the generalizability of our results. However, Dallas is a populous, metropolitan area, and our jail population of PLWH is likely similar to that of other US cities.

Our findings from this unique dataset, which integrates criminal justice and community clinic data, have important implications for STI screening and secondary HIV prevention. We found low numbers of testing for GC/CT, particularly in the jail, which we attributed to testing protocols. We identified a high proportion of person with syphilis, compared to both the general U.S. population and PLWH. We observed that a disproportionate number of syphilis cases were found in black individuals and MSM. This occurred in the setting of a high proportion with detectable HIV VL concurrent with a diagnosis of a new STI, concerning for possible HIV transmission. We propose that routine, opt-out screening for STIs, including rectal and oropharyngeal screening, conducted both during incarceration and after release, has the potential to identify high numbers of asymptomatic infections. This approach to STI testing provides not only a platform to diagnose and treat STIs in a high risk population with limited access to care, but also a chance for safer sex counseling, STI prevention, and ART adherence counseling.