Incarcerated persons are disproportionately affected by sexually transmitted infections (STIs), particularly Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC), with several studies reporting higher prevalence among young incarcerated women.1–3 Furthermore, STIs diagnosed within correctional facilities represent a substantial proportion of overall reported cases in several urban cities.4,5 A substantial number of infections may be asymptomatic and, if untreated, may result in morbidity such as salpingitis, ectopic pregnancy, pelvic inflammatory disease, and infertility.6 The presence of ulcerative and nonulcerative STIs has additionally been linked to a 3- to 5-fold increase in the risk of HIV acquisition,7 an important consideration in this high-risk population. Correctional facilities offer a unique opportunity to provide screening, treatment, and prevention services to an assembly of at-risk persons who may not routinely seek health care. In the case of transmissible and curable infections, treatment within correctional facilities may benefit not only the individual but also the larger public health. If untreated, detainees released to their community may proceed to transmit STIs to uninfected sexual partners, amplifying the cases in the community.
Efforts in screening and treatment have several challenges. The patient population within jails is transitory and constantly in flux, with most detainees released or transferred within 1 week. Effective protocols for screening and treatment must therefore be rapid processes. The development of nucleic acid amplification tests that simplify specimen collection (urine sample compared with urethral or cervical swab) and can provide rapid results have greatly improved the feasibility for GC/CT screening in jails. Additional challenges such as designation and availability of personnel to administer screening, logistics of obtaining specimen at intake, and the ever-persistent budgetary limitations require a collaborative approach between correctional staff and health care workers to maximize both security and health promotion.8 Given the potential benefit in this population, the Centers for Disease Control and Prevention support recommendations for universal GC/CT screening for incarcerated adolescents and women younger than 35 years, although most jails offer testing only at the request of the inmate rather than universal screening.9,10 When implemented, screening methods vary considerably among correctional facilities in terms of targeted populations and whether screening is performed for those who accept the offer (“opt-in”) versus automatic screening with the option to decline screening (“opt-out”). Opt-out screening has several potential advantages such as increasing the proportion of population screened by capturing asymptomatic persons and reducing any stigma or discrimination associated with testing based on inmate request.
The Cook County Jail (CCJ) located in Chicago, IL, is one of the largest single-site county jails in the United States with an annual census of more than 100,000 persons. From 2008 to 2011, testing for chlamydia, gonorrhea, HIV, and syphilis was not routine and was performed based on inmate request, reported symptoms, or risk factors. With political support from the 10th District Cook County Commissioner, the government provided funding to initiate universal opt-out screening. Program design and monitoring plan was accomplished in collaboration with the Chicago public health department, Department of Justice, CCJ medical providers, and electronic medical record technical staff. Aside from an additional phlebotomist hired to support increased screening (for HIV and syphilis), no additional human resources were required; instead, the current staff who were already trained in STI screening were trained in a new protocol and script. In April 2011, opt-out screening for chlamydia and gonorrhea was initiated among women at the time of intake. This target population was chosen for initial roll-out based on the high prevalence of GC and CT in this group and the smaller population size in which to implement the new process. The objectives of this study were to evaluate the impact of opt-out screening on rates of testing and diagnosis of gonococcal and chlamydial infections in incarcerated women, assess the proportion of infections successfully treated, and evaluate factors associated with receipt of treatment.
MATERIALS AND METHODS
Participants and Data Collection
We conducted a retrospective review of the results of opt-out screening for GC/CT of all female detainees from April 2011 to December 2012. At intake to CCJ, all detainees undergo a medical evaluation administered by Cermak Health Services medical personnel. For women, an automatic electronic order is generated for urinary GC, CT, and pregnancy tests. Intake screening includes a written general medical consent, posted signage in multiple languages notifying women of the opt-out screening policy, and a verbal script providing instructions on the opt-out process. Written declination of some or all of the tests is documented, at which point the automatically generated order is canceled by intake staff and test status recorded as “canceled.” Test status was recorded as “incomplete” if testing was not performed for any reason when a detainee had not opted out (such as failure to obtain specimen or specimen mishandling). The order status was used as a surrogate for opt-out status as the written opt-out form for every patient was not reliably available for this review. Urine was tested for gonorrhea and chlamydia by nucleic acid amplification test (BD ProbeTecTM 133 ET, Franklin Lakes, NJ). Specimen processing was performed off-site within 24 hours at the John H. Stroger Hospital of Cook County laboratory. Positive results were reported to Cermak Health Services staff, and antibiotic treatment was administered under observation; chlamydial infection was treated with oral azithromycin, and gonococcal infection was treated with oral cefixime until August 2012, when treatment was changed to intramuscular ceftriaxone according to updated Centers for Disease Control and Prevention guidelines. Patients with gonococcal infection received empiric therapy for chlamydia. Treatment was documented in the electronic medical record. Test of cure was not routinely performed, except in the case of pregnancy. All positive results were also reported to the Chicago Department of Public Health (CDPH). Women of childbearing age (<35 years) who did not receive treatment in jail were followed up by CDPH after release, and treatment was attempted. The study was approved by the institutional review boards at Cook County Health and Hospital Systems and Rush University Medical Center.
We collected the status (completed, incomplete, and canceled) of GC/CT electronic orders and results of all GC/CT tests ordered at intake for women from April 2011 through December 2012. To assess the impact of the initiation of opt-out screening, we collected the results of any GC/CT testing performed in women before opt-out screening (January 2011 through March 2011). Chart review was performed for all women with a positive result to assess whether treatment was administered; time to treatment, defined as number of days between when results were reported and when treatment was given; and length of detention.
Associations between patient characteristics and GC or CT infection and receipt of treatment were assessed using χ2 tests for categorical variables and Wilcoxon rank sum tests for nonnormally distributed continuous variables. Trends in testing and positivity over time were assessed by nonparametric trend test. Differences in average time to treatment by day of intake were compared using the Kruskal-Wallis test. We ran logistic regression models to examine univariable and multivariable associations between patient characteristics (age, pregnancy status), length of stay in jail, and intake on weekend versus weekday with receipt of treatment. Factors with P < 0.2 in univariable analysis were entered in multivariable models; odds ratios from the multivariable models represent the independent association of each variable with receipt of treatment, adjusted for the other factors in the model. Age was collected as a continuous variable and categorized for analysis as <24, 24–30, and >30 years to aid in the interpretation of odds ratio and prevalence estimates. Length of stay was dichotomized about the median (10.9 days) for analysis because of a high degree of skewness in the variable distribution. Model fit was assessed with the Hosmer-Lemeshow goodness-of-fit test and was considered adequate based on nonstatistically significant χ2 test for lack of fit. Analyses were conducted with SAS version 9.2 (SAS Institute, Cary NC).
A total of 17,065 women were eligible for opt-out screening from April 2011 through December 2012. The median age was 32 (range, 16–79) years. Urine nucleic acid amplification test was performed in 9265 (54.2%) of eligible detainees. The average turnaround time for test results was 1.75 days. Overall, 22% (3729 women) of eligible detainees opted out of screening. Compared with women who declined testing or whose tests were incomplete, women who were successfully tested were somewhat younger (median age, 31 vs. 32; P < 0.001). Over time, the proportion of detainees that opted out decreased from an average of 28.3% in 2011 to 16.8% in 2012 (P < 0.001). There were 4071 tests (24%) that were characterized as incomplete with a significant increase over time from an average of 10.1% in 2011 to 34% in 2012 (P < 0.001). Corresponding to the increase in incomplete tests, the rate of screening declined during the study period from 68% 180 of eligible women screened during April to June 2011 to 44.6% screened during October in December 2012 (P < 0.001).
Of the 9265 women screened during the study period for whom results were available, 235 (2.5%) gonococcal and 702 (7.6%) chlamydial infections were diagnosed; combined prevalence of either STI was 9.4% (866/9255). Sixty-two (7.2%) were pregnant at the time of STI diagnosis. GC and CT infections were significantly more common among younger women (Table 1). There was no significant change in the prevalence of GC or CT during the study period (P = 0.51 and P = 0.17, respectively). Of the 866 detainees with positive test results, 69.5% (602/867) received treatment during incarceration; treatment rates remained constant during the study period (P = 0.58). Older age, pregnancy, and longer duration of incarceration were associated with higher likelihood of treatment, whereas screening performed on a weekend was associated with lower likelihood of treatment (Table 2). In multivariable analysis, age 24 to 30 years versus <24 years (adjusted odds ratio [aOR], 1.73; 95% confidence interval [CI], 1.14–2.63), pregnancy (aOR, 2.51; 95% CI, 1.22–5.18), and longer length of stay in jail (aOR, 18.1; 95% CI, 11.7–28.1) were associated with greater odds of treatment (Table 3). The association between intake to the jail on a weekend versus weekday and receipt of treatment was not significant after controlling for length of stay. Median length of incarceration in those who were treated was 19.3 (8.8–57.7) days, compared with 2.3 (1.5–3.9) days in untreated detainees. The average duration from intake to treatment was 2.1 days. The mean time to treatment was shorter for pregnant women compared with nonpregnant women (1.5 days vs. 2.1 days, P = 0.002; Table 4). Intake later in the week was associated with longer duration to treatment (P < 0.001). There were no differences in time to treatment between age groups (P = 0.09).
The average number of monthly GC and/or CT tests was 155 per month before the initiation of opt-out screening versus 455 tests per month after opt-out screening, a 2.8-fold increase. The average number of diagnoses in the pre–opt-out period was 9.3 cases/mo compared with 40.8 cases/mo after initiation of opt-out screening. The jail census of female detainees was similar between time periods.
The results of this study demonstrate the high prevalence of bacterial STIs in detained women (9.4%). Similar to prior reports, chlamydia was more common than gonorrhea, and prevalence was inversely proportional to age.1,2,11 Implementation of opt-out screening successfully increased the average number of diagnoses per month by 4.4-fold compared with the number of cases per month in the pre–opt-out period when testing was performed based on inmate request. Because a high proportion of chlamydial infections among women are asymptomatic, routine screening is important for interrupting transmission and preventing serious sequelae, particularly among high-risk women. In this study, expanding from symptom-based testing to opt-out screening increased the case detection by 4.4-fold, suggesting that a significant proportion may have been asymptomatic infections.
Identifying and treating asymptomatic infections may prevent amplification of infections in the community when infected detainees are released. Broad et al.12 demonstrated this association of community and jail-based STI diagnoses. After routine GC/CT screening was discontinued among male detainees at the CCJ in Chicago, IL, the reported cases of GC and CT in men citywide decreased by 33% and 20%, respectively, related to a significant decline in case finding and treatment; during the same period, diagnoses of CT in women citywide increased 2.5%, suggesting an increase in transmission. Barry et al.13 demonstrated the potential benefit of jail-based screening and treatment for overall community rates of CT infections. Screening and treatment of young incarcerated adults were associated with a 50% reduction in community-diagnosed chlamydial infections in neighborhoods that have high incarceration rates.
Opt-out screening was performed at intake, in conjunction with routine medical screening. A minority (22%) of women in this study chose to opt out of screening. Previous studies have evaluated opt-out HIV screening in men and women in jail and report higher rates of opting out (50% in women) and suggest that when screening is offered within the first 24 hours, it is more likely to be accepted.14,15 In comparison with HIV screening, which may have more social and psychological complexities, bacterial STI screening may be more acceptable; reports of opt-in GC/CT screening have demonstrated similarly high acceptance rates.16 The reasons for a decrease in the proportion of women who opted out over time are unclear. Recidivism is common, and it is possible that women repeatedly detained developed increased tolerance of the standard procedure. In a brief survey conducted after the initiation of opt-out screening, women identified a variety of reasons for opting out, including self-identified low-risk, previously tested by community health provider, and incomplete understanding of the process.17 Further studies to characterize the women who routinely opt out of STI screening would be worthwhile. Although the percentage of women opting out declined over time, the rate of screening did not significantly increase accordingly. This is likely explained by the increase in the proportion of incomplete tests. There are several reasons that a test may be recorded as incomplete: if the woman is transferred or released from intake before she provides a specimen, if she is unable to provide a specimen, if the specimen is not transferred to the lab, and laboratory mishandling. It is also possible that there were women who opted out of screening without a cancellation of the electronic order, leading to an overestimate of incomplete tests. There was no clear change in protocol or processing during this period to provide a single explanation. Given a prevalence of 9.4%, an estimated 382.7 cases, or 18.2 cases/mo, were missed due to incomplete screening. This finding illustrates the need for continual efforts in collaboration between health care and correctional staff in the implementation of these large-scale efforts, as decline in interest and attention may lead to decay in the rate of screening and decreased effectiveness of the intervention.
The ultimate goal of increased screening efforts is to provide effective treatment and thereby prevent long-term complications. This is particularly important in correctional settings where likelihood of postrelease follow-up is low, and information available to Public Health officials for the sake of tracking may be incomplete and/or falsified. Administering therapy in jails can be difficult. Intake is a standardized process that applies to each detainee and thus represents an optimal point to apply screening tests. However after intake, when results are available, the geographic placement within the jail, court schedules, and duration of incarceration for each detainee are highly variable. This variability creates several barriers to administration of therapy. At CCJ, 70% of women with GC/CT were successfully treated while in jail. Of the 264 detainees that were not treated in jail, 220 (83%) were younger than 35 years and would therefore be followed up by CDPH, who would attempt to track and administer therapy within the community; the success rate for this population in the community is currently unknown. With few jails providing universal screening programs, there are limited data on treatment rates, which have ranged from 63% to 75% of in-jail treatment.10 In a review of 3 large jails in which GC/CT screening was offered to a proportion of detainees, treatment was successfully administered in jail in 35% to 72% of cases.16 In a voluntary screening program in San Francisco, a significantly higher proportion of adolescents received treatment, and in total, 78% of all infections were treated in jail.18 Models suggest that in high-prevalence settings with high rates of treatment, routine jail-based screening for GC/CT is cost-effective compared with syndromic management based on the significant costs related to the potential complications of untreated infections in women.19
We hypothesized that many women would be released before they could receive treatment. Indeed, the median length of incarceration in those who were not treated was significantly shorter (2.3 days vs. 19.3 days, P < 0.001). In general, treatment was administered promptly with a median of 1 day (range, 0–28) days, suggesting that factors in addition to length of stay may be involved in receipt of treatment. Weekends at CCJ notably have reduced medical staff available, and we therefore examined whether the day of the week that the test was performed was related to receipt of treatment. Although detainees tested later in the week had a longer duration to treatment in univariable analysis, this was no longer significant when adjusting for length of stay in multivariable analysis. Interventions aimed at improving the process of STI result reporting and treatment delivery, particularly for women with a short detention, should be evaluated with further studies.
There are limitations to our study. Our prevalence estimates are based on the number of women screened and not the entire female population entering jail. The women who were not screened were slightly older than those who did receive screening, and it is possible that there were other unmeasured differences between these groups. Opt-out status was captured in electronic medical record review as canceled in order status. If women opted out in writing, but the test was not manually canceled, these would have been recorded as incomplete test, thus artificially lowering the numbers of women who opted out and artificially increasing the number of incomplete tests. Although no change in protocol was instituted among correctional staff, we hypothesize that this may have contributed to the increase in incomplete tests over time but was not measureable in this study. About successfully treating cases of GC/CT, we did not examine the number of women who may have opted in and requested testing later in their detention. Also, owing to available resources, we were unable to obtain details on community treatment in women who were not treated at the jail. If significant numbers of women were found and treated in the community, it would further support the use of increased screening in the jail population for the purposes of reducing the overall burden of disease.
Bacterial STIs are an important medical and public health concern in correctional facilities, and management is confronted with several challenges. Our report shares the experience of implementing opt-out screening for GC/CT of women in a large urban jail. Opt-out screening effectively increased the number of women screened and cases diagnosed, and the jail successfully treated most infections. Effective large-scale screening and treatment programs require a dedicated and continuous collaboration between correctional staff, health care providers, and public health departments.
1. Joesoef MR, Weinstock HS, Kent CK, et al. Sex and age correlates of chlamydia prevalence in adolescents and adults entering correctional facilities, 2005: Implications for screening policy. Sex Transm Dis 2009; 36 (2 suppl): S67–S71.
2. Mertz KJ, Voigt RA, Hutchins K, et al. Findings from STD screening of adolescents and adults entering corrections facilities: implications for STD control strategies. Sex Transm Dis 2002; 29: 834–839.
4. McIntyre AF, Studzinski A, Beidinger HA, et al. HIV/AIDS, and hepatitis services in Illinois County Jails. Sex Transm Dis 2009; 36 (2 suppl): S37–40.
5. Pathela P, Hennessy RR, Blank S, et al. The contribution of a urine-based jail screening program to citywide male chlamydia and gonorrhea case rates in New York City. Sex Transm Dis 2009; 36 (2 suppl): S58–61.
6. Stamm W, Batteiger B. Chlamydia trachomatis
(trachoma, perinatal infections, lymphogranuloma venerum, and other genital infections). In: G.L. Mandell J.E. Bennett, Dollin R. (eds.). Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 7th ed. Geneva, Switzerland: Churchill Livingstone; 2009: 2443–2461.
7. Wasserheit JN. Epidemiological synergy. Interrelationships between human immunodeficiency virus infection and other sexually transmitted diseases. Sex Transm Dis 1992; 19: 61–77.
8. Arriola KR, Braithwaite RL, Kennedy S, et al. A collaborative effort to enhance HIV/STI screening in five county jails. Public Health Rep 2001; 116: 520–529.
9. Center for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2010. MMWR Morb Mortal Wkly Rep 2010; 59: 11–12.
11. Hammett TM. Sexually transmitted diseases and incarceration. Curr Opin Infect Dis 2009; 22: 77–81.
12. Broad J, Cox T, Rodriguez S, et al. The impact of discontinuation of male STD screening services at a large urban county jail: Chicago, 2002–2004. Sex Transm Dis 2009; 36 (2 suppl): S49–52.
13. Barry PM, Kent CK, Scott KC, et al. Is jail screening associated with a decrease in chlamydia positivity among females seeking health services at community clinics? San Francisco, 1997–2004. Sex Transm Dis 2009; 36 (2 suppl): S22–28.
14. Kavasery R, Maru DS-R, Cornman-Homonoff J, et al. Routine opt-out HIV testing strategies in a female jail setting: A prospective controlled trial. PloS One 2009; 4: e7648.
15. Kavasery R, Maru DS-R, Sylla LN, et al. A prospective controlled trial of routine opt-out HIV testing in a men’s jail. PLoS One 2009; 4.
16. Mertz KJ, Schwebke JR, Gaydos CA, et al. Screening women in jails for chlamydial and gonococcal infection using urine tests: Feasibility, acceptability, prevalence, and treatment rates. Sex Transm Dis 2002; 29: 271–276.
17. Zawitz C, Livak B, Owens F, Grasso A. HIV/STI opt-out testing prevalence rates among females after re-implementation of opt-out testing at one of the largest correctional facilities in the United States. Presented at: 19th International AIDS Conference (THPE570), Washington DC, 2012.
18. Barry PM, Kent CK, Scott KC, et al. Optimising sexually transmitted infection screening in correctional facilities: San Francisco, 2003–2005. Sex Transm Infect 2007; 83: 416–418.
© Copyright 2014 American Sexually Transmitted Diseases Association
19. Kraut-Becher JR, Gift TL, Haddix AC, et al. Cost-effectiveness of universal screening for chlamydia and gonorrhea in US jails. J Urban Health Bull N Y Acad Med 2004; 81: 453–471.