Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Ann Emerg Med. Author manuscript; available in PMC 2016 July 1.
Published in final edited form as:
PMCID: PMC4478148

HIV Care Continuum for HIV-Infected Emergency Department (ED) Patients in an Inner-City Academic ED

Yu-Hsiang Hsieh, PhD,1 Gabor D. Kelen, MD,1 Oliver Laeyendecker, PhD,2,3 Chadd K. Kraus, DO, MPH,4 Thomas C. Quinn, MD, MSc,2,3 and Richard E. Rothman, MD, PhD1,2



The recently released “HIV Care Continuum” is a cornerstone of the National AIDS Strategy and a model for improving care for those living with HIV. There are no studies exploring the entirety of the HIV Care Continuum for patients in the emergency department (ED).


To determine gaps in the HIV Care Continuum in order to identify potential opportunities for improved care for HIV-infected ED patients.


A mixed methods approach was employed in one inner-city ED in 2007. Data elements were derived from (1) An identity-unlinked HIV seroprevalence study; (2) An ongoing non-targeted HIV screening program; and (3) A structured survey of known HIV-positive ED patients.


Identity-unlinked testing of 3,405 unique ED patients found 265 (7.8%) were HIV positive. Of those testing HIV positive, 73% were previously diagnosed [based on self-report, chart review and/or presence of antiretroviral (ARVs) in serum], but only 61% were recognized by the clinician as HIV infected (based on self-report or chart review). Of those positive, 43% were linked to care, 39% were retained in care, 27% were on ARVs, 26% were aware of their receiving ARV treatment, 22% were virally suppressed, and only 9% were self-aware of their viral suppression.


This study is the first to quantify gaps in HIV care for an ED patient population using the HIV Care Continuum as a framework. Our findings identified distinct phases (i.e. testing, “provider awareness of HIV diagnosis”, and linkage to care) where the greatest opportunities for intervention exist, if appropriate resources were allocated. This schema could serve as a model for other indolent treatable diseases frequently seen in EDs, where continuity of care is critical.

Keywords: HIV, emergency department, undiagnosed infection, linkage to care


HIV transmission is directly dependent on HIV viral load.1,2 Viral suppression can be effectively accomplished via antiretroviral therapy (ART).3 Achievement of undetectable viral load is recommended as the ultimate goal for HIV prevention and control.2 There are five steps in the ‘HIV Care Continuum’ required for reaching undetectable viral loads for those infected with HIV, these are 1) initial HIV diagnosis, 2) linkage to care, 3) retention in care, 4) receipt of ART, and 5) achieving a low viral load. The importance of this conceptual framework was first quantified by Gardner et al in a landmark study which depicted the full spectrum of HIV care across the United States, reporting that only 19% of the population reach that target.4 Notably, that study employed data from diverse patient populations, varied geographic regions, and multiple years, yielding a relatively simplistic model for defining the gaps in at each step along the continuum.5 Based on that schema, President Obama issued an executive order in 2013, coined The HIV Care Continuum Initiative, which was directed towards accelerating improvements in HIV prevention and care across the United States.6,7 Since then, many investigators have applied this framework in different settings to identify gaps along the continuum, in order to assess where, when and what resources could best be applied to improve HIV care.8-13 To date, there have been no studies evaluating the HIV Care Continuum from an emergency department (ED) perspective.

For more than a decade, EDs have been playing a critical role in HIV testing efforts,14,15 with thousands of previously undiagnosed HIV-infected individuals identified (principally via dedicated HIV testing programs).16 Since EDs also function as a safety net for more than a half a million individuals living with HIV17 many of whom use the ED as their only source of care, potential opportunities for impacting other aspects of HIV care exist. ED-based HIV testing initiatives have already been shown to be effective for identifying those with undiagnosed HIV and linking newly diagnosed patients to primary or HIV specialty care. In practice emergency clinicians also routinely establish or re-establish care for patients with known HIV infection, which frequently includes ensuring that patients are receiving appropriate ARV therapy. The Ability to reliably assess the clinical status of ED patients with HIV (both recognized and unrecognized), in relation to the stated goals of the HIV Care Continuum is important, given that the ED is frequently the primary or sole point of entry into the health care system for those patients.18 Gaps in provider knowledge such as recognition of a patient's HIV serostatus, ARV adherence and stage of disease, may impact management decisions for patients with HIV. Reasons for these knowledge gaps are varied but include limitations in ability to test patient's in some EDs, relative lack of availability of clinical information, and variable patients self-reporting regarding their HIV diagnosis, treatment, and extent of viral suppression.

The objective of this study is to determine gaps in the HIV Care Continuum for patients with HIV who visit the ED in order to identify potential opportunities for improved care, and to target aspects of our own screening program where potential opportunities for improvement exist. Establishing a framework for an ED HIV Care Continuum can also serve as a prototype for investigators who are studying other indolent chronic infectious or non-infectious diseases that are frequently undiagnosed and/or undertreated. The data for informing our ED Continuum of Care model was derived from a single urban academic ED using multiple data sources from a single calendar year, permitting a more reliable assessment of this environment. We also propose and highlight three new operationally defined stages along the Care Continuum which are relevant to ED settings, these are 1) “provider awareness of HIV diagnosis”, 2) “patient on ARV treatment – patient self-aware”, and 3) “viral load suppression– patient self-aware”. These proposed additional ‘stages’ are particularly relevant to settings where the clinicians and/or patient may have incomplete information related to the patient's disease state, and have downstream consequences for both individual and population based control of HIV/AIDS.

Materials & Methods

The studies were conducted at an academic, inner-city adult ED with 60,000 visits (in 2007) that primarily serves a local socio-economically disadvantaged population in Baltimore City, Maryland. Our ED has a high prevalence of HIV infection, and has nearly two decades of experience in HIV testing, treatment, and follow-up.19-24 Beginning in 2005, we instituted an ED-based rapid HIV screening program.25

During the summer and fall of 2007, we conducted three simultaneous HIV studies in our ED: (1) an identity-unlinked HIV seroprevalence study; (2) a programmatic evaluation of an established ED-based rapid HIV screening program; and (3) a cross-sectional survey study of known HIV-positive patients designed to assess their experience with HIV care. A mixed method approach was used to develop the estimates for each stage of the HIV Care Continuum. All three studies were approved by the Johns Hopkins University School of Medicine Institutional Review Board.

Study 1, was an HIV identity-unlinked seroprevalence study conducted from June 23, 2007 to August 18, 2007 which included all ED patients ≥ 18 years who had blood drawn for clinical procedures. During the study period, there were 9,179 ED visits. Among them, there were 4,475 for whom unused excess blood specimens were collected. The identity-unlinked seroprevalence study methodology has been well described in the literature19 and involved collection of excess sera as part of clinical procedures, assigning of a unique study code, and removal of all identifiers and protected health information from samples following collection of basic data (e.g. age; gender; race; risk factors). For each patient from whom blood was obtained, the blood sample and the data collected were labeled with unique code numbers. The data collected excluded all forms of patient identification and associated protected health information. Thus, no result could be traced back to any specific patient and protected health information was not available in the study database. Chart reviews for study 1 were undertaken in a specific, structured manner using generally suggested methods for chart review that have been widely published in the emergency medicine literature.26 These methods were used in previously published HIV research in this ED.19,20 Five trained research assistants participated in the study design and development stages of the project. The design of the standardized data abstraction format was based on our prior published studies.19,20,24 An electronic data abstraction form was created with hard stops and parameters intended to maximize data accuracy (e.g. age parameters; binary response choices where appropriate). Structured data forms defined specific variables, which were extracted from the ED chart and the electronic medical record (EMR). The research staff received training in clinical research methods, chart review, and laboratory methods prior to collecting any data. Staff were supervised throughout the design and collection period by a senior research staff member with formal graduate level research training. A senior clinical research staff member with graduate-level research training did spot-checking of the data collected by each research assistant. No formal inter-rater reliability assessment was performed. Weekly study meetings were held to address specific data collection issues such as discrepancies in data available in the chart, from the patient, and in the EMR. Discrepancies were adjudicated by consensus of the senior faculty investigator on the study team. Regarding used clinical blood samples, de-identified samples were tested for HIV by third generation HIV enzyme immunoassays (EIA); positive EIAs were confirmed by Western blot followed by RNA viral load testing using Roche Amplicor v1.5 which has a limit of detection of 400 copies/ml (Roche, Indianapolis, IN). Known HIV positivity was determined by either chart review or self-report as part of the HIV screening program, and collected prior to data de-identification. Antiretrovirals (ARVs) in serum specimens were detected using ultra-performance liquid chromatography-tandem mass spectrometry by the Clinical Pharmacology Laboratory at the Johns Hopkins University.27 These tests detect the presence of ARV in the serum specimens among the vast majority of individuals on ART. Specifically, the following analytes of ARV were tested: Amprenavir (APV), Atazanavir (ATV), Darunavir (DRV), Efavirenz (EFV), Indinavir (IDV), Lopinavir (LPV), metabolite of nelfinavir (M8), Nelfinavir (NFV), Nevirapine (NVP), Ritonavir (RTV), Saquinavir (SQV), and Tipranavir (TPV). Thus, all the preferred and alternative regimens of ART were tested except for AZT/3TC/ABC which was not a recommended regimen in 2007. Results of this HIV seroprevalence study have not been previously reported in the peer-reviewed literature.

Study 2, was a non-targeted, rapid HIV screening program initiated in this ED in 2005, and used the OraQuick Advance Test (OraSure Technologies, Bethlehem, PA). During July and August 2007, an exogenous staffing model (i.e. HIV testing facilitator based) was employed. Thirteen trained testing facilitators, working in shifts offered HIV testing to eligible ED patients, 24 hours a day, 7 days a week. Facilitators, who were not part of the medical treatment team, consented patients, performed brief pre-test counseling, collected oral specimens, disclosed test results, and performed post-test counseling. Rapid tests were run in the on-site ED clinical laboratory by laboratory staff. A dedicated HIV program coordinator was responsible for linkage to care for any newly diagnosed HIV-infected patients. During this period of time, 1,173 ED patients were tested for HIV. Seven newly diagnosed cases were identified and four (57%) patients were successfully linked to care. From August 2006 to June 2007 and September 2007 to December 2007, a hybrid staffing model in which the ED-based rapid HIV screening program was staffed which included both exogenous facilitators, and indigenous medical staff (based on time of day and day of week). During this period time, 1,332 ED patients were tested, 27 were newly diagnosed with HIV infection and 16 (59%) were successfully linked to care. Together, from August 2006 to December 2007, 31 newly diagnosed HIV-infected patients were identified from our ED HIV screening program and 20 (59%) patients were linked to care. Details of the testing program and a subset of the linkage to care data have been described and reported elsewhere.25

Study 3 was a structured survey conducted in the same ED during July and December of 2007, of known HIV-positive ED patients. That investigation was designed to assess various aspects of patient's impression of HIV care from self-report, including frequency of scheduled HIV care in the past 12 months, receipt of ARV medication, and most recent range of viral loads. One hundred and seven patients were enrolled; details of this survey have been previously described.28 Briefly, known HIV-positive ED patients were eligible for this pilot study if they were not critically ill or capable of providing informed consent. At the time of enrollment, participants completed an interviewer-administered structured questionnaire about their demographic, behavioral, and clinical characteristics, as well as health care information including age, race, sex, highest level of education achieved, employment status, income level, housing condition, sexuality, injection drug use, types of medical insurance, HIV-related health care information, length of HIV diagnosis, antiretroviral drug use, and comorbid conditions. Most recent HIV viral load information (undetectable; detectable but <5,000; 5,000-99,999; ≥ 100,000; or unknown/unsure) was also collected via patient self-report in the questionnaire. HIV ARV treatment and viral load information used as data source for this study have not been reported in the literature.28

Key data elements from the three studies/programs described above were used to estimate the numbers for each stage of the HIV Care Continuum for our ED in 2007. We used total number of HIV-infected ED patients identified in the seroprevalence study (Study 1) as the first stage of the HIV Care Continuum, and then estimated the remaining numbers in step-wise progression. First, we estimated the number of patients who had previously diagnosed HIV infection, defined based on either chart documented and/or self-reported HIV-positive status, and/or the presence of ARV in the specimen from the seroprevalence study. Next, we used the proportion of successful linkage to care from our 2007 screening program (Study 2), to estimate the numbers of HIV positive patients linked to care. Then, we estimated the retention in care number by applying the proportion estimated from the survey study (Study 3). We next estimated the number of patients in treatment (based on presence of ARV in serum) and virally suppressed (from direct viral load data). We also estimated provider and patient awareness of the stages along the HIV Care Continuum. An aware HIV diagnosis (with awareness defined from the perspective of the ED provider) was determined using data from the seroprevalence study (infection was chart documented) or from the screening program (patient self-reported positivity to testing facilitator staff). The number of patient self-aware regarding ARV treatment was determined based on the proportion of known HIV positive patients in care who reported that they were on ARV treatment (Study 3), using the number of patients retained in care as the denominator. Similarly, the number of patient self-aware that they were virally suppressed was determined by the proportion of known positive patients in care who reported their viral loads to be at an undetectable level (Study 3), using the number of patients retained in care as the denominator. The 95% confidence interval (CI) of each data element from the three studies/program described above was calculated. A sensitivity analysis was performed to provide the upper and lower bounds of each estimate for each stage of the HIV Care Continuum using upper and lower limits of the 95% CI for each data point from three data sources.

Operational definitions which were derived from the literature or authors' consensus were as follows: An HIV infection was defined as a reactive HIV EIA test confirmed by Western blot. A diagnosed infection was defined as either chart-documented HIV infection, self-reported infection from the screening program, and/or the presence of ARV in the serum specimen. “Provider awareness of HIV diagnosis” was defined as either chart-documented HIV infection, or self-reported infection from the screening program, indicating the treating provider was likely aware of the patient's HIV serostatus. Self-awareness of ARV treatment and self-awareness of undetectable viral loads were taken from self-reported information from the known HIV-positive survey described above. Linked to care was considered unsuccessful if there was no documented evidence of a patient with confirmed positive results entering into care within 6 months of the initial reactive ED rapid test, despite two attempts at follow-up by the program coordinator.25,29,30Retention in care was defined as at least two scheduled clinic visits for HIV care in the past 12 months, as reported by survey participants from Study 3.28,31,32


The estimated numbers for each stage of the HIV Care Continuum from this ED in 2007 are presented in Figure 1. From the identity unlinked seroprevalence study (Study 1), 265 (7.8%) of 3,405 unique subjects presenting to the ED were identified as HIV-infected. Of these 192 (72.5%; 95% CI: 66.8%, 77.6%) were considered diagnosed infections: 162 (61.1%; 95% CI: 55.2%, 66.9%) were either chart documented or self-reported as HIV positive (i.e., provider awareness of HIV diagnosis); an additional 30 (11.3%) patients whose HIV positive status was not documented or reported to the facilitator staff, had ARVs detected in their serum sample.

Figure 1
HIV Care Continuum with Introduction of Three New Operationally Defined Stages for HIV-Infected Emergency Department (ED) Patients in an Inner-City Adult Emergency Department in 2007

Applying the observed 59% (95% CI: 41.9%, 74.3%) successful linkage to care proportions for newly diagnosed infections from our 2007 HIV screening program, we estimated that 113 (43%) patients of the 192 diagnosed HIV positive patients were linked to care. Next, applying an observed retention to care proportion of 64.5% (95% CI: 55.1%, 73.1%) of the total 162 patients who were documented or self-reported to be HIV positive (Study 3 data, HIV-positive ED patient survey), we estimated that 104 of 265 (39%) patients were retained in care, or 92% of those patients linked to care, were retained in care.

Our laboratory data found 71 of 265 (27%, 95% CI: 22%, 32%) HIV positive patients were currently on ARV treatment. Taken together with the survey data which found that 66.7% (95% CI: 55.0%, 77.0%) of known HIV positives patient in care report currently being on ARV treatment, we estimated that 69 of 265 (26%) of HIV positive patients were self-aware of their treatment. Finally, our ARV testing data demonstrated that 57 of 265 (22 %; 95% CI: 17%, 27%) patients were virally suppressed, yet only 23 (9%) patients were estimated to be self-aware of being virally suppressed. This latter estimate is based on survey data indicating that 21.7% (95% CI: 13.2%, 32.6%) of known positives who reported being in care, understood that they had undetectable viral loads.

Regarding the three proposed new operationally defined steps along the HIV Care Continuum, notably a substantial proportion of patients in the ED who were infected with HIV were found to have gaps (communication and/or knowledge), related to their care status. That is, in the step “provider awareness of HIV diagnosis”, 15.6% (30/192) of providers were unaware their patient was HIV infected, and in the step of “viral load suppression– patient self-aware”, 59.6% (34/57) of patients were unaware that they were virally suppressed. For the step of “patient on ARV treatment – patient self-aware”, all but two patients (97.2%) were self-aware of their ARV treatment status.

The upper and lower bounds of our estimate for each step of HIV Care Continuum by sensitivity analysis are presented in Figure 1. Among all 265 HIV-infected ED patients, 67% to 78% were diagnosed (i.e. 22% to 33% of all infections were undiagnosed), 55% to 67% were diagnosed and aware to provider, 30% to 54% were linked to care, 30% to 49% were retained in care, 22% to 32% were on ARV treatment, 17% to 37% were self-aware of ARV treatment, 17% to 27% were viral suppressed, and 4 to 16% were self-aware of viral suppression.


This study has several limitations. First, this is a cross-sectional snapshot of the HIV Care Continuum in one adult ED with a high seroprevalence of HIV in the population. Most U.S. EDs have a lower prevalence of infection. Additionally, the ED population in our study has had a dedicated ED-based HIV testing program in place since 2005, with feasibility demonstration programs dating back to 1995.33 Therefore, our results might not be generalizable to EDs with a lower HIV seroprevalence, or those with less dedicated or sustained ED-based HIV testing resources. Still, because our ED has been one of the nation's epicenters for conducting ED-based epidemiological and clinical studies of HIV,19,23,25,28,33 our findings do provide important proof-of-concept data regarding the potential role that EDs provide in addressing the HIV Care Continuum. The framework presented could also serve as a prototype for other investigators and/or policy makers involved in ED-based efforts for HIV testing and linkage to care. Second, similar to the original HIV Care Continuum described by Gardner and colleagues,4 our care continuum estimates were not derived from longitudinal data from a cohort of HIV-infected patients. Instead, the data we present are from three studies conducted in parallel in a single ED. The simultaneous conduct of the studies (all in a single year) makes it likely that the three study populations came from a common and relatively fixed or static population;34 this makes these estimates even more reliable than those used to model the original HIV Care Continuum, which were drawn from multiple populations in different geographical regions and settings over a span of more than 10 years.4 Third, it is possible that were some patients who should have been labeled as having diagnosed HIV infection, but were not categorized appropriately. Instances where this may have occurred include those in whom (1) the patient's HIV positive status was not documented in the medical charts at our institution; and/or (2) the patient chose not to disclose their positive status to our HIV screening program staff; and/or (3) the HIV-infected patient was not taking ARV (i.e. absence of ARVs in their serum specimens). Therefore, our number of undiagnosed infection could be over-estimated. Fourth, translating our estimates of the HIV Care Continuum for our ED patients should be made with caution, since the data were gathered in 2007. Advances occurred since 2007 which could impact these estimates include: improvement in HIV testing diagnostics, namely use of fourth generation HIV EIA assay which shortens the window period; improvements of our ED screening program which has increased the proportions of those offered and tested; introduction of a more active linkage to care protocol which include use of dedicated case management; active involvement of our affiliated HIV clinic staff which has yielded improved retention in care and adherence to ARV treatment; more aggressive ARV treatment guidance; and enhanced efficacy of ARV regimens for suppressing HIV viral loads. Taken in combination these factors likely would produce a different picture of the metrics shown here for an ED centric HIV Care Continuum, than the one described herein. Still, these 2007 estimates provide a useful framework for understanding factors which contribute to the HIV Care Continuum, as well as a baseline set of comparator data for making improvement to our HIV screening program. Finally, our estimates of the HIV Care Continuum in our ED are potentially limited by biases inherent to individual data sources including: selection bias associated with laboratory specimens (e.g. which patients had blood drawn and available for the identity-unlinked seroprevalence study); demographic/clinical data obtained from chart review (e.g. absence of systematic documentation of HIV positivity on patient's chart and absence of information regarding HIV care outside of our institution); and information biases from our surveys (e.g. HIV serostatus and self-report regarding treatment and viral suppression in the absence of confirmation from chart review). The magnitude of these selection and information biases cannot easily be estimated. Specifically, our HIV prevalence estimates were made from patients who were 18 years and older and had blood drawn as part of their ED care. Accordingly it is possible that the true HIV prevalence could be higher or lower, but that would be purely speculative. Thus, we were not able to adjust for selection and information biases derived from individual data sources.


The study is the first, as far as we are aware to define gaps in the spectrum of HIV care and the associated missed opportunities along each stage of the HIV Care Continuum for an ED patient population. Compared to estimated figures described in the original national HIV Care Continuum, which used multiple data sources from the late 1990s to 2008,4 our 2007 ED data found that HIV-infected patients who visited this ED had slightly higher estimated proportions of undiagnosed infection (27% vs. 21%), lower proportions of linkage to care (43% vs. 59%), but similar proportions of retention in care (39% vs. 40%), ARV treatment (27% vs. 24%), and viral suppression (22% vs. 19%). These findings provide a framework and a starting set of data for future ED investigators conducting HIV related research to consider.

The HIV Care Continuum has been widely described for internal medicine/infectious disease and public health audiences, but is a relatively new concept for emergency medicine. In the past decade, there has been a large investment including dedicated public health resources, and focused ED investigation directed towards optimizing ED-based HIV testing, referral, and linkage to care processes. Some in the ED community considered this investment as a laudable goal35,36 since many ED patients may not get tested elsewhere, and early identification of infected patients is beneficial for future patient management and treatment.37 However, as highlighted in the original HIV Care Continuum schema,4 testing and linkage to care are only two stages of the HIV care cascade. Although few question the important role EDs are already playing in initial HIV diagnosis and early linkage to care, our role in attending to downstream stages along the care continuum has never been addressed.

What is the impact EDs could potentially play in both individual HIV illness and community HIV transmission, were we to engage in other aspects of the HIV Care Continuum such as re-linking those who have fallen out of care,38 encouraging retention in care, and/or counseling adherence to ARV treatment? Conceptually this introduces discussion, and likely some controversy, regarding the appropriate balance between providing emergent/urgent care for ED patients and providing continuity of care services. An important consideration in that equation is recognition that such downstream services (for example, adequate viral suppression) are tied to future ED and inpatient resource utilization, metrics which are receiving increased national attention. Multiple other questions must be considered in applying these findings to practice however, such as what are the external resources (e.g. funding, manpower, and infrastructure) required to support such activities, and from where will they be drawn? In this regard, even the existence of the most basic ED-based HIV testing programs has been hotly debated in the literature39,40 since the time that EDs were first called out as one of the key health care sites for testing in the 2006 CDC recommendations for HIV testing.41 While compelling data support some form of ED-based screening,42 it remains to be determined what approach is optimal for individual EDs, and whether such efforts are truly sustainable for the long term.15,43 Thus, while the conceptual framework of the HIV cascade may make sense from a broad clinical and public health perspective, feasibility studies are needed, and ultimately expert consensus required for developing pragmatic guidance regarding what role, if any, EDs can or should play at each stage along the HIV Continuum Care.

Identifying undiagnosed HIV infection, i.e. testing, is the first step in the cascade of the HIV Care Continuum, and the most critical for achieving high levels of viral suppression in the population (i.e. viral suppression will be limited if there are significant number of undiagnosed individuals, even if there were no or little gaps along the remaining steps of the cascade).4 It is noteworthy, that in spite of the presence of our long standing ED HIV screening program25 in which 0.5 ~ 2% of ED patients tested for HIV were newly diagnosed infection, a substantial proportion of HIV infections remained unrecognized. Specifically, we found 27% of HIV infections in all HIV-infected patients were previously undiagnosed. Such a high proportion of undiagnosed infections likely results at least in part, from the high prevalence of HIV (11-12% in 2001-2003) in the community our ED serves. This drives HIV transmission and is reflected in the observed HIV incidence of 0.5 to 1% per year.24 Another possible explanation for the high prevalence of undiagnosed infections reported here, is that implementation and uptake of HIV screening was not sufficiently comprehensive at the time the study was conducted (2007). That possibility is supported by findings from a separate study from that same year, where we discovered 10 fold higher prevalence of undiagnosed HIV amongst ED patient who were not offered testing, and 5 fold higher prevalence of undiagnosed cases amongst patients who were offered but declined testing, versus those who accepted testing (where the prevalence was 0.4%).44 Still another possible explanation for the high prevalence of undiagnosed infections observed in our ED, is that some individuals who we considered ‘undiagnosed infections’ did in fact have acute HIV infections, which we were not able to recognize using the OralQuick rapid HIV test platform, which has a longer window period than the newer 4th generation HIV testing technologies.45 Finally, it is possible that some cases we considered ‘undiagnosed infections’ were actually previously diagnosed but inappropriately categorized, due to either nondisclosure of serostatus by the patient and/or inadequate chart documentation.

Successfully linking a newly diagnosed HIV-infected patient to care is a complicated and multi-factorial process, which can be challenged by both patient and system-level barriers. The proportion of our patients who were linked to care at the time of this study (approximately 60% of those newly diagnosed) is relatively lower than that which has been observed more recently, as reported in a meta-analysis which found that 76% of HIV-positive patients tested from EDs and urgent care settings were successfully linked.30 The most likely explanation for the relatively low proportion of patients who were linked to care (in 2007) was the fact that we employed a relatively passive approach to follow-up care, which consisted of calling the clinic to arrange the next available appointment and up to 2 follow-up calls after missed appointments. Since then, our ED, and most others have resorted to more active linkage to care methods, which include partnerships with case management, social work and on site specialty care clinics, often involving patients being taken directly from the ED to their first intake visit.46,47 Such programs, as with ours, have yielded significant improvements in linkage to care with ours achieving over 90% success. Interestingly, the proportion of patients linked to care from varied community settings (including sexually transmitted diseases clinics, primary care clinics, community-based organizations, health departments, homeless shelters, single-room occupancy hotels, and other unspecified community HIV testing settings) was reported in that same meta-analysis to be significantly lower (69%) than that seen in EDs and urgent care settings.30 Investigators who conducted the meta-analysis study speculated that the availability of onsite HIV clinics within the same institution was the primary driver for that improved outcome. If their conjecture is true, EDs might well be optimally positioned to help narrow this critical gap in the national HIV Care Continuum. Providing the appropriate follow-up and linkage to care for urgent conditions is in fact keeping with one of the primary roles EDs play for all of our patients.

In addition to effectively linking newly diagnosed HIV-infected ED patients to care, EDs could play a role in re-linking known HIV-positive patients to care – including those who were never in care, as well as those who fell out of care. The potential value of that intervention was first proposed by Lyons et al38 who demonstrated both patient acceptability and feasibility (albeit, when supplementary resources were employed) for such an intervention. Conceptually, EDs could leverage fixed infrastructure put in place for linking to care those with newly diagnosed HIV (e.g. use of case managers, or streamlined referral methods), permitting sustainable programs for re-linking or promoting retention in care for those with known HIV infection. Again, however, the value and cost of offering those services need to be carefully evaluated, within the framework of the ED mission.

For the HIV Care Continuum, we propose consideration of three new stages which are particularly relevant for ED populations in this study- namely “provider awareness of HIV diagnosis”, “patient on ARV treatment – patient self-aware”, and “viral load suppression– patient self-aware”. Provider awareness of HIV diagnosis is well recognized to be important for ED clinical management decisions, and has been previously reported to impact clinical decision-making.48,49 Patient self-awareness of being on ARV therapy may be relevant for ED management decisions (including referral for treatment, potential medication related complications, and/or contraindications for medications prescribed in the ED). Patient self-awareness regarding whether they are virally suppressed or not, may be critical for forming appropriate differentials based on the patient's ED presenting complaint. Furthermore, since ARV treatment is now recommended for all HIV positive individuals, regardless of CD4 count (both for reducing risk of disease progression and prevention of transmission of HIV in the community),50,51 patient awareness of their viral suppression status could help clinicians optimize timing of linkage to care as appropriate. From a public health standpoint, introducing the concept of patient self-awareness regarding their viral suppression status, could be a marker for slowing community transmission (similar to what was previously found by regarding patient self-awareness of their HIV serostatus),52 though further investigation would be required to demonstrate that. Finally, these additional proposed ‘awareness’ stages along the HIV Care Continuum, could serve as targets for HIV surveillance efforts, as well as assessment tools for intervention programs.

Finally, the HIV Care Continuum framework provides a potential model for discussion about the role EDs play in the management of other treatable chronic diseases. There are a wide variety of underdiagnosed and/or untreated indolent chronic diseases we see amongst our ED population, e.g. hypertension, diabetics, hepatitis C, and depression, where advancing a disease-specific care continuum model could yield improved outcomes, were effective holistic strategies considered. Advancement of the HIV Care Continuum grew out of the sometimes fragmented care which emerged with the evolving epidemic and treatment strategies, providing opportunities for improved care of this population, many of which remain untapped as highlighted by the numerous gaps which remain.4 Regardless of whether or not EDs engage in taking on the challenge of the gaps in the continuum of care, consideration of the framework for clinical and public health outcomes are relevant to the emergency medicine community.

In conclusion, this study represents the first introduction of the HIV Care Continuum framework to emergency medicine professionals and provides a set of estimates for the gaps in the care continuum from an ED patient population perspective. Our findings highlight the presence of multiple potential ‘missed’ opportunities along the continuum of HIV care for ED patients, but leave questions regarding how the ED or others can effectively manage these gaps.


The study was supported by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health. The Johns Hopkins University Department of Emergency Medicine HIV Testing Program was funded in part by grants from Maryland Department of Health and Mental Hygiene. Dr. Hsieh is also supported in part by an NIH Award, 5K01AI100681 from NIAID to study HIV testing in emergency departments using a modeling approach.


1. Quinn T, Wawer M, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med. 2000;342:921–9. [PubMed]
2. Cohen M, Smith M, Muessig K, Hallett T, Powers K, Kashuba A. Antiretroviral treatment of HIV-1 prevents transmission of HIV-1: where do we go from here? Lancet. 2013;382:1515–24. [PMC free article] [PubMed]
3. Cohen M, Chen Y, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493–505. [PMC free article] [PubMed]
4. Gardner E, McLees M, Steiner J, Del Rio C, Burman W. The spectrum of engagement in HIV care and its relevance to test-and-treat strategies for prevention of HIV infection. Clin Infect Dis. 2011;52:793–800. [PMC free article] [PubMed]
5. Das M, Chu P, Santos G, et al. Decreases in community viral load are accompanied by reductions in new HIV infections in San Francisco. PLoS One. 2010;5:e11068. [PMC free article] [PubMed]
6. Executive Order -- HIV Care Continuum Initiative - Accelerating improvements in HIV prevention and care in the United States through the HIV Care Continuum Initiative. [Accessed August 20, 2014];The White House Office of the Press Secretary. 2013 at
7. HIV/AIDS Care Continuum. U.S. Department of Health & Human Services, 2013. [Accessed January 26, 2014]; at
8. Whiteside Y, Cohen S, Bradley H, et al. Progress along the continuum of HIV care among blacks with diagnosed HIV- United States, 2010. MMWR Morb Mortal Wkly Rep. 2014;63:85–9. [PubMed]
9. Birger R, Hallett T, Sinha A, Grenfell B, Hodder S. Modeling the impact of interventions along the HIV continuum of care in Newark, New Jersey. Clin Infect Dis. 2014;58:274–84. [PMC free article] [PubMed]
10. Hall H, Frazier E, Rhodes P, et al. Differences in human immunodeficiency virus care and treatment among subpopulations in the United States. JAMA Intern Med. 2013;173:1337–44. [PubMed]
11. Axelrad J, Mimiaga M, Grasso C, Mayer K. Trends in the spectrum of engagement in HIV care and subsequent clinical outcomes among men who have sex with men (MSM) at a Boston community health center. AIDS Patient Care STDS. 2013;27:287–96. [PMC free article] [PubMed]
12. Dombrowski J, Kent J, Buskin S, Stekler J, Golden M. Population-based metrics for the timing of HIV diagnosis, engagement in HIV care, and virologic suppression. AIDS. 2012;26:77–86. [PMC free article] [PubMed]
13. Nosyk B, Montaner J, Colley G, et al. The cascade of HIV care in British Columbia, Canada, 1996-2011: a population-based retrospective cohort study. Lancet Infect Dis. 2014;14:40–9. [PMC free article] [PubMed]
14. Haukoos J. The impact of nontargeted HIV screening in emergency departments and the ongoing need for targeted strategies. Arch Intern Med. 2012;172:20–2. [PubMed]
15. Hsieh YH, Wilbur L, Rothman R. HIV Testing in U.S. Emergency Departments: At the Crossroads. Acad Emerg Med. 2012;19:975–7. [PubMed]
16. Centers for Disease Control and Prevention. Results of the Expanded HIV Testing Initiative--25 jurisdictions, United States, 2007-2010. MMWR Morb Mortal Wkly Rep. 2011;60:805–10. [PubMed]
17. Mohareb A, Rothman R, Hsieh Y. Emergency department (ED) utilization by HIV-infected ED patients in the United States in 2009 and 2010 - a national estimation. HIV Med. 2013;14:605–13. [PubMed]
18. Kelen G, Rothman R. Emergency department-based HIV testing: too little, but not too late. Ann Emerg Med. 2009;54:65–71. [PubMed]
19. Kelen G, Fritz S, Qaqish B, et al. Unrecognized human immunodeficiency virus infection in emergency department patients. N Engl J Med. 1988;318:1645–50. [PubMed]
20. Kelen G, Hexter D, Hansen K, Tang N, Pretorius S, Quinn T. Trends in human immunodeficiency virus (HIV) infection among a patient population of an inner-city emergency department: implications for emergency department-based screening programs for HIV infection. Clin Infect Dis. 1995;21:867–75. [PubMed]
21. Kelen G, Shahan J, Quinn T. Emergency department-based HIV screening and counseling: experience with rapid and standard serologic testing. Ann Emerg Med. 1999;33:147–55. [PubMed]
22. Kelen G, Shahan J, Rothman R, Ketlogetswe K, Quinn T. Long-term trends in HIV infection in an academic medical center ED. Acad Emerg Med. 2002;9:368–9.
23. Kraus C, Rothman R, Shahan J, Quinn T, Kelen G. Trends in HIV infection in the ED: a 16 year review. 2006 Society for Academic Emergency Medicine Annual Meeting; San Francisco, CA. 2006. 2006.
24. Laeyendecker O, Rothman R, Henson C, et al. The effect of viral suppression on cross-sectional incidence testing in the johns hopkins hospital emergency department. J Acquir Immune Defic Syndr. 2008;48:211–5. [PMC free article] [PubMed]
25. Hsieh YH, Jung J, Shahan J, et al. Outcomes and Cost Analysis of Three Operational Models for Rapid HIV Testing Services in an Academic Inner-City Emergency Department. Ann Emerg Med. 2011;58:S133–9. [PubMed]
26. Gilbert E, Lowenstein S, Koziol-McLain J, Barta D, Steiner J. Chart reviews in emergency medicine research: Where are the methods? Ann Emerg Med. 1996;27:305–8. [PubMed]
27. Avery L, Parsons T, Meyers D, Hubbard W. A highly sensitive ultra performance liquid chromatography-tandem mass spectrometric (UPLC-MS/MS) technique for quantitation of protein free and bound efavirenz (EFV) in human seminal and blood plasma. J Chromatogr B Analyt Technol Biomed Life Sci. 2010;878:3217–24. [PMC free article] [PubMed]
28. Soong T, Jung J, Kelen G, et al. Is inadequate human immunodeficiency virus care associated with increased ED and hospital utilization? A prospective study in human immunodeficiency virus-positive ED patients. Am J Emerg Med. 2012;30:1466–73. [PubMed]
29. Lyons M, Lindsell C, Haukoos J, et al. Nomenclature and definitions for emergency department human immunodeficiency virus (HIV) testing: report from the 2007 conference of the National Emergency Department HIV Testing Consortium. Acad Emerg Med. 2009;16:168–77. [PMC free article] [PubMed]
30. Marks G, Gardner L, Craw J, Crepaz N. Entry and retention in medical care among HIV-diagnosed persons: a meta-analysis. AIDS. 2010;24:2665–78. [PubMed]
31. Sherer R, Stieglitz K, Narra J, et al. HIV multidisciplinary teams work: support services improve access to and retention in HIV primary care. AIDS Care. 2002;14:S31–44. [PubMed]
32. Horstmann E, Brown J, Islam F, Buck J, Agins B. Retaining HIV-infected patients in care: Where are we? Where do we go from here? Clin Infect Dis. 2010;50:752–61. [PubMed]
33. Kelen G, Hexter D, Hansen K, et al. Feasibility of an emergency department-based, risk-targeted voluntary HIV screening program. Ann Emerg Med. 1996;27:687–92. [PubMed]
34. Fertel B, Hart K, Lindsell C, Ryan R, Lyons M. Toward understanding the difference between using patients or encounters in the accounting of emergency department utilization. Ann Emerg Med. 2012;20:693–8. [PMC free article] [PubMed]
35. Haukoos J, Mehta S, Harvey L, Calderon Y, Rothman R. Research priorities for human immunodeficiency virus and sexually transmitted infections surveillance, screening, and intervention in emergency departments: consensus-based recommendations. Acad Emerg Med. 2009;16:1096–102. [PMC free article] [PubMed]
36. Haukoos J, Hopkins E. Understanding HIV screening in the emergency department: is perception reality? Acad Emerg Med. 2013;20:309–12. [PMC free article] [PubMed]
37. Transitions of Care Task Force Report. American College of Emergency Physicians. [Accessed September 10, 2014];2012 at
38. Lyons M, Raab D, Lindsell C, Trott A, Fichtenbaum C. A novel emergency department based prevention intervention program for people living with HIV: evaluation of early experiences. BMC Health Serv Res. 2007;7:164. [PMC free article] [PubMed]
39. Irvin C, Flagel B, Fox J. The emergency department is not the ideal place for routine HIV testing. Ann Emerg Med. 2007;49:722. [PubMed]
40. McKenna M. HIV testing: should the emergency department take part? Ann Emerg Med. 2007;49:190–2. [PubMed]
41. Branson B, Handsfield H, Lampe M, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep. 2006;55:1–17. [PubMed]
42. Holtgrave D. Assessing the Population-Level Prevention Effect of Emergency Department-Based HIV Testing in the United States: A Research Framework and Commentary. Ann Emerg Med. 2011;58:S164–7. [PubMed]
43. Moran G, Talan D. Processes and Models for HIV Screening in the Emergency Department: Can and Should We Do This? Ann Emerg Med. 2011;58:S172–3. [PubMed]
44. Hsieh YH, Beck K, Kelen G, et al. STI & AIDS World Congress. Vienna, Austria: 2013. High Prevalence of Undiagnosed HIV Infection in Patients Who Were Not Offered Screening and Patients Who Declined Screening: Evaluation of a Rapid HIV Screening Program in a U.S. Urban Emergency Department. 2013.
45. Rothman R, Saheed M, Hsieh Y. Infectious disease/CDC update Detection of acute HIV infection in two evaluations of a new HIV diagnostic testing algorithm--United States, 2011-2013. Ann Emerg Med. 2014;63:56–60. [PMC free article] [PubMed]
46. Leider J, Fettig J, Calderon Y. Engaging HIV-positive individuals in specialized care from an urban emergency department. AIDS Patient Care STDS. 2011;25:89–93. [PMC free article] [PubMed]
47. Menon A, Martis M, Nganga-Good C, Rothman R, Hsieh Y. Linkage to Care Methods and Rates in U.S. Emergency Department-Based HIV Testing Programs - A Systematic Review and Meta-Analysis. SAEM 2014 Annual Meeting; Dallas, TX. 2014.
48. Lubelchek R, Kroc K, Hota B, et al. The role of rapid vs conventional human immunodeficiency virus testing for inpatients: effects on quality of care. Arch Intern Med. 2005;165:1956–60. [PubMed]
49. Lubelchek R, Kroc K, Levine D, Beavis K, Roberts R. Routine, Rapid HIV Testing of Medicine Service Admissions in the Emergency Department. Ann Emerg Med. 2011;58:S65–70. [PubMed]
50. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services. [Accessed September 3, 2014];2012 at
51. Thompson M, Aberg J, Hoy J, et al. Antiretroviral treatment of adult HIV infection: 2012 recommendations of the International Antiviral Society-USA panel. JAMA. 2012;308:387–402. [PubMed]
52. Marks G, Crepaz N, Janssen R. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS. 2006;20:1447–50. [PubMed]