In 2010, the White House released the first comprehensive National HIV/AIDS Strategy. [1] Since the publication of this Strategy, data released by the Centers for Disease Control and Prevention show that there are significant gaps along the continuum of HIV care—the sequential stages of care from diagnosis to viral suppression. [2] Nearly one in six of the estimated 1.1 million people living with HIV in the United States are unaware of their infection and an estimated 48,000 new infections occur each year.[3] Implementing HIV screening in a cost efficient manner in venues with large numbers of persons who may not otherwise be tested is an important step in reducing the number of persons unaware of their infection and improving linkage to subsequent medical care.

Emergency departments (EDs) and jails provide medical services to persons at risk for HIV infection and are recommended venues for HIV screening,[4,5] but typically less than 1% of persons entering either venue are tested for HIV.[6,7] Previous studies have highlighted the cost-effectiveness of HIV testing in EDs and jails and the importance of identifying newly diagnosed HIV cases.[8,9] Despite the CDC guidelines for routine HIV screening in healthcare settings [5] and CDC guidance to correctional facilities on incorporating HIV testing into routine medical services,[4] surveys have shown fewer than a quarter of EDs in the U.S. offer systematic HIV testing programs, and the largest jails in most communities where HIV prevalence is high do not routinely provide HIV screening during the intake medical evaluation.[6,7]

Safety-net hospitals and jails located in the inner-cities often manage similar populations and provide health care to persons experiencing health disparities, including those who are underinsured, low socio-economic status, and racial and ethnic minorities. Studies conducted in Georgia have shown neighborhoods disproportionately affected by poverty have a higher than average percentage of residents with a history of incarceration.[10] Not only are populations of the two institutions shared, but the processes by which people are evaluated for medical services in the venues are similar. Nationally, an ED in a public, safety-net hospital has a median length of stay of 5.0–7.5 hours.[11] A jail is a short-term correctional facility, operated by a local municipality or county, where the median length of stay is 2 to 5 days.[12,13] In both venues, rapid throughput of entrants is an important goal. Staff may express concerns that adding routine HIV screening could overburden the organization’s ability to meet that goal. Nonetheless, evidence has shown that incorporating HIV testing into the operations of both EDs and jails is feasible.[14,15] Cost data on ED and jail-based HIV screening programs are necessary for program administrators considering providing routine HIV screening in these settings.

Grady Memorial Hospital (GMH) and the Fulton County Jail (FCJ) are located approximately 4.5 miles apart in the downtown area of Atlanta, Georgia. At the end of 2010, 18.2% of county residents were living below the poverty level and 1.3% of Fulton County residents were infected with HIV.[16,17] Previous studies have documented that HIV screening at both GMH and the FCJ venues identify a high number of HIV infections, including infections among men who have sex with men and racial minorities.[18–20] This paper describes the cost of HIV screening in the Grady Memorial Hospital Emergency Department (GMHED), which sees primarily adult patients, and the Fulton County Jail, and reports the cost per newly diagnosed case of HIV infection detected in each of the two screening programs.

In June 2008, the GMHED initiated a rapid HIV screening program utilizing a parallel staffing model whereby HIV testing staff offer testing separate from the regular ED medical services.[18,19] The program continued through December 2011, for a total of 40 months. The program was staffed by three part-time counselors, one part-time database manager, one full-time program coordinator and an ED physician with funded time (0.15 FTE) for program supervision. Patients were identified by the HIV testing staff as ineligible for HIV screening if they were less than 18 years of age, known to have HIV/AIDS, non-English speaking, incarcerated, medically unstable or unable to decline testing. The HIV counselors approached eligible patients from 10:00am to 10:00pm Monday through Friday in either the ED waiting room or in individual examination rooms. The trained HIV counselors reviewed the triage list and approached eligible patients in consecutive order, and explained that rapid HIV screening was part of routine medical care. In accordance with hospital requirements, patients were given the opportunity to decline the HIV screening test and were asked to sign a form indicating their decision to accept or decline. Patients accepting HIV screening received pre-test prevention counseling. GMHED’s HIV testing data were collected prospectively for quality assurance and analyzed retrospectively for research purposes. The retrospective analysis was approved by Emory University’s Institutional Review Board (IRB) and Grady Memorial Hospital’s Research Oversight Committee.

In January 2011, nursing staff at the FCJ began integrated routine rapid HIV screening into the medical intake process, which occurs immediately after booking. Data were analyzed through December 2011 (excluding 6 weeks midsummer 2011, when the contractor providing medical services for FCJ changed). Nurses provided HIV screening 24 hours a day, 7 days per week during medical intake evaluations. Prior to intake, nursing staff obtained written consent from the detainees for general medical care, including HIV screening. Nurses informed the detainee-patient that voluntary HIV screening was a routine screening test conducted during the intake medical evaluation and the patient had the right to decline the test. The extensiveness of HIV prevention counseling provided was based on the clinical judgment of the nurse. HIV risk information was obtained for all persons with a reactive HIV test result. All detainees with a reactive rapid test result were referred for additional HIV medical evaluation. Additionally, an HIV educator provided on-going HIV training to the nursing staff and visited newly diagnosed persons within one day of the preliminary diagnosis to supplement counseling provided by the nurses. HIV-positive patients or those determined by nursing staff to be mentally incompetent were not screened for HIV. The protocol for this program was reviewed by Emory University’s IRB and determined that it was public health practice rather than human subjects’ research.

The GHMED had approximately 360,000 patient visits during the 40-month study period. 19,388 eligible patients were approached: no risk behavior assessment or determination of signs or symptoms was conducted before offering HIV screening. Of those that were offered the test, 15,510 (80.0%) were tested (7,562 men, 7,946 women and 2 male-to-female transgender persons). Two hundred and sixty-eight oral fluid tests were preliminarily positive. Project staff confirmed that 26 persons had been previously diagnosed. The remaining 242 (90.3%) were confirmed positive with Western blot. These 242 remaining were classified as newly diagnosed persons with HIV infection (1.56% sero-positivity). (Table 1)

During the 11-month study period at FCJ, 30,799 detainees were provided an intake medical evaluation and thus were eligible for HIV screening. While delays in training new staff, staffing shortages, and unavailability of test kits occurred periodically, 18,183 FCJ entrants actually had the opportunity to undergo HIV screening. HIV screening occurred on 11,819 (65.0%) detainees who were approached (9,467 men and 2,352 women); the remaining 35.0% declined an HIV test. One hundred and thirty tests were preliminarily positive and potentially represented new diagnoses. However, of the 130 persons with positive oral fluid test results, 89 were reclassified as previously diagnosed and 41 (31.5% of preliminary positive entrants) were confirmed as newly diagnosed cases of HIV infection (0.35% sero-positivity). (Table 1)

For the purpose of cost analyses, comprehensive program costs were used to calculate cost per patient and per newly diagnosed case of HIV infection. Costs of test kits ($12.50 per kit) were $193,875 (includes cost of donated kits) in the GMHED and $147,737 in the FCJ. The costs of running controls were $8,970 in the GMHED and $8,000 in the FCJ. Confirmatory Western blots were performed at total costs of $10,300 and $5,356 in the GMHED and the FCJ respectively. In terms of labor, for the GMHED testing program, the total labor cost for the 40-month period was $508,218, of which approximately $381,000 represented the labor of counseling and testing. Overall personnel cost for the FCJ testing program was $113,132 for the 46 weeks of testing, and included salary and fringe benefits for nurses ($38,132, calculated at $20/hour for work force primarily composed of licensed practical nurses) and an HIV educator ($75,000). On average, nurses spent 9.67 (S.D. 4.87) minutes per each test performed; during the time tests were developing, nurses could perform other duties, such as recording vital signs, and obtaining the prior medical history of a detainee-patient, thus test processing time was excluded in our cost analysis.

The GMHED’s total program cost was $721,363 with an average cost per test of $46.51. In the FCJ, the total cost of the program was $274,225, and the mean cost for each test was $23.20. The mean cost per new HIV diagnosis was $2,981 in the GMHED and $6,688 in the FCJ.

The HIV screening programs implemented in both the GMHED and FCJ were able to identify HIV-positive persons who were previously unaware of their infection. A recent review showed that cost per new diagnosis in screening programs ranged from $2,000 to $30,000.[24] The costs per new diagnosis in both of these settings are lower than comparable HIV testing cost analyses performed elsewhere. The cost per new HIV diagnosis may vary in these screening programs by the differences in methods for confirming a new HIV diagnosis, previously undiagnosed HIV infection, and the methods used to determine program costs. The GMHED cost for each newly identified HIV-positive diagnosis was $2,981 versus a median nationwide of $20,144 (2009 US$) for costs analyzed by program budget methods.[22] The jail’s cost of $6,688 per confirmed new HIV positive test was also lower compared to the median, nationwide figure of $15,018 for costs derived by microcosting-staff allocation methods. [22]

While neither program was able to offer screening to all potential patients, FCJ's integrated testing model utilizing jail nursing staff permitted broader testing coverage and resulted in lower costs per test. The GHMED’s parallel staffing model resulted in limited HIV screening due to lack of 24-hour availability of HIV test counselors. However, the high HIV test acceptance rate in the ED was encouraging. Implementation of an integrated HIV screening model in the GMHED would increase the availability of HIV testing services and likely increase the number of newly diagnosed cases of HIV infection.

Testing in each of these venues reaches patients who are at high risk of infection. Blacks are the population group most affected by HIV in the United States. Despite representing only 12% of the US population, in 2010, black men and women accounted for 44% of all new HIV infections for persons 13 years or older. Specifically, black men accounted for 70% of these new infections and 72% of those men identified as gay or bisexual, or other men who have sex with men (MSM).[25]

Previously published analyses from the GMHED testing program demonstrated that a vast majority of persons testing positive were black (95.2%) and male (75.2%),[19] and a disproportionate number of persons testing positive without a prior positive test in the GMH system were MSM (16.1%) compared to heterosexual male patients (1.5%).[18] Analysis of the FCJ data has also shown that 38% of men newly diagnosed with HIV are black men who have sex with men.[20] Given that nearly half (48.5%) of all health care provided to young black men is delivered through emergency departments [26] and one in five black men are incarcerated by age 34 years,[27] HIV screening in both emergency departments and jail sites in high HIV prevalence communities represents important opportunities to reach populations disproportionately affected by HIV in an efficient manner.

HIV screening in both the emergency department and jail venues yielded a high number of HIV diagnoses at low cost per new diagnosis. Different methods for confirming a new HIV diagnosis, previously undiagnosed HIV sero-positivity, and determination of program costs, prevent conclusions from being drawn that one location or testing program is more economical than the other.