Differences Between Proof-of-Concept Studies and Effective Implementation: Routine, Opt-Out HIV Testing in Emergency Departments

Rotheram-Borus, Mary Jane PhD; Klosinski, Lee E PhD; Etzel, Mark A MPP

JAIDS Journal of Acquired Immune Deficiency Syndromes: 1 December 2007 - Volume 46 - Issue 4 - pp 381-383
doi: 10.1097/QAI.0b013e3181582da9

From the Center for HIV Identification, Prevention, and Treatment Services, University of California, Los Angeles, Los Angeles, CA.

Correspondence to: Mary Jane Rotheram-Borus, PhD, Center for Community Health, 10920 Wilshire Blvd., Suite #350, Los Angeles, CA 90024-6521 (e-mail: cchpublications@mednet.ucla.edu).

Article Outline

In 2006, the Centers for Disease Control and Prevention (CDC) announced guidelines recommending routine HIV testing in all health care settings anticipated to have an HIV prevalence rate of >0.1% among its population.1 Between 1% and 4% of the patients served in emergency departments (EDs) are anticipated to be HIV seropositive.2,3 Given the CDC policy, HIV testing programs have recently been evaluated in 6 EDs,2-5 including the article by Brown3 in this journal. Overall, the studies demonstrate that routine, opt-out HIV testing is feasible in EDs, resulting in cost-effective identification of HIV-positive patients and the ability to link HIV-positive patients to care. Together, these studies prove the concept of using ED-based testing to help to identify the 25% of HIV-positive adults unaware of their serostatus and the 40,000 annual incident HIV infections in the United States.1

However, there were substantial variations in each of the EDs in terms of the percentage of patients who were approached for HIV testing (2.1% to 47.5%), those who accepted testing (52% to 98.3%), and the percentage of HIV-positive patients linked to care after testing (50% to 90.8%).2-5 The desired goal would be for ED staff to routinely approach most patients for HIV testing, for almost all patients approached to agree to be tested, and then for those who are identified as HIV-positive to be linked to care. Yet none of the proof-of-concept studies came close to meeting this goal. Together, these studies suggest that there are significant opportunities to improve the rates of offering, accepting, testing, and linking patients to care when moving from research studies to actual implementation in EDs nationwide. This article outlines the opportunities to improve the rates from those observed in the proof-of-concept research studies. There are both policy and organizational changes that must accompany broad diffusion of the CDC's guidelines for HIV testing.8

Researchers are allowed into EDs at the pleasure of the hospital administrators, staff, and patients. Researchers' goals are typically to disturb an agency as little as possible while trying to test their program or theory. To implement a sustainable organizational change, such as routinely implementing HIV testing, requires structural changes and shifts in staff roles, rules, and interpersonal relationships. For example, when implementing new procedures (like ED HIV testing and counseling within an organization), there are at least 4 different processes that must be in place before implementation to ensure high adoption: 1) Administrative support at the highest levels. If administrators are not supportive of an innovation, it is guaranteed that there will be no consequences for staff failing to implement the innovation. Change is never desirable for its own sake, and administrators' support is critical for securing staff investment in change. 2) Written and broadly known procedures for implementing the innovation. Before initiating any innovation, brainstorming by administrative and line staff is a good strategy to identify the optimal procedural guidelines to maximize patient adherence to the innovation and to minimize the costs of initiating new procedures (time, inconvenience, dealing with unpleasant staff, increased funding). Procedures that are integrated with existing, routine tasks (eg, completing intake forms) are the easiest to introduce and to sustain over time with fidelity. In a research study, the project is time-limited, so it is seldom integrated into ongoing routines. For example, in the Brown study, undergraduate volunteers approached each patient, rather than completing HIV testing by having an existing professional staff member ask 1 additional question.3 3) Staff training. All staff must be aware of the innovation, even if the task (ie, HIV testing) is not the direct responsibility of the specific staff member. Recursive quality mechanisms get established in well-functioning organizations. All staff must know the tasks that must be completed with each patient to complete HIV testing and linkage to care; in these organizations, all staff can spontaneously ensure that all procedures are completed with each patient. Staff training must be targeted and received by all staff in all 3 shifts in a hospital (overnight, morning, afternoon/evening shifts). In a research project, the investigators' goal of minimizing the research burden results in nonoptimal organizational policies and less staff training. 4) Continuous quality improvement procedures. To achieve continuous quality improvement, a method for monitoring and promoting staff compliance when implementing the HIV testing program is necessary. Effective monitoring helps to identify areas for improvement. Information about how well the program works can be its own reward; therefore, establishing a monitoring system is basic to its success. In 1 research project in an ED, we sent a letter to the personnel file for each staff member who helped to ensure that a patient adhered to treatment after their hospital visit; the personnel letter rewarded staff for following the protocol. The specific rewards within an agency can be quite diverse (eg, donuts in the morning break room, a $5 coupon to the hospital book store, an award at the monthly staff meeting), but must be considered to enhance staff adherence to a procedure.

There are multiple variations in state HIV testing policies, some of which are likely to maximize acceptance of HIV testing and linkage to care, and other policies that inhibit testing. For example, 14 states currently require written consent or pretest counseling before conducting HIV testing.6 Sites requiring written informed consent report lower rates of acceptance compared to sites not requiring written consent or pretest counseling.

A second policy is the national rapid testing algorithm used. Some countries utilize 2 positive rapid tests to diagnose HIV infection and require a third confirmatory rapid test when the first 2 tests results are discrepant: 2 positive tests indicate HIV-positive status, 2 negative tests indicate HIV-negative status, and discrepant results indicate the need for confirmatory testing.7 United States guidelines for rapid testing dictate that initially reactive (or positive) rapid tests be confirmed with a Western blot test.5 The required Western blot confirmatory test often creates a significant delay, usually of several days or weeks, until the results of the HIV testing are available for those testing HIV-positive on the rapid test. In the Brown study,3 there were 4 of 14 false positives using only the 1 rapid test; in each of these 4 cases, a second rapid test would likely have identified the false positive. Delays in receiving the results of the Western blot test occurred repeatedly across the 6 ED studies: when test results were delayed, a substantial number of patients were lost to follow-up and not informed of their HIV-positive serostatus.2-5

Is it time to reconsider a change in United States policy and deliver an HIV diagnosis based on the results of 2 matching rapid tests? This change in testing algorithm would permit immediate feedback on a patient's HIV status and eliminate almost all of the loss to follow-up of HIV-positive patients. Enzyme immunoassay and Western blotting would still need to be used to confirm weak positive bands on rapid tests and as a quality assurance measure mechanism.7 The proposed change in rapid testing algorithm would not completely eliminate the need for delays associated with confirmatory serological testing. It would, however, reduce the number of patients in this status and promote immediate linkage to care for those determined to be positive based on 2 positive rapid tests. As treatment for HIV infection improves dramatically, adjudicating the dual responsibilities of providing scientifically confident test results and linking HIV-positive patients with ongoing care may lead us to rethink national testing policy.

Finally, if an ED is not reimbursed for its procedures, it will be very difficult to sustain the action. An effective national ED HIV testing strategy will only be sustainable with reasonable reimbursement for HIV testing.5-7,9 Currently, there are no funding streams that are specified for routine, opt-out HIV testing.

Research studies mounted to demonstrate proof-of-concept do not provide positive models of how to implement sustainable programs. It would be desirable that in the design of future interventions, researchers systematically address how the policy and organizational environments influence the observed outcomes. In the case of HIV testing in EDs, staff integrated into the ongoing clinic operation and endorsement from the hospital administrators and line supervisory staff are needed before beginning to implement a program. Providing staff with the necessary and relevant training and implementing continuous quality improvement procedures are essential to maximizing the uptake and sustainability of programs like HIV testing in EDs. Rather than being an additional task that requires verbal patient consent, testing that is integrated into routine care procedures is desirable, with some additional financial reimbursement attached.

Today, the gap between research on HIV testing in the ED and real-world implementation of testing programs is relatively large. Future researchers must close this gap by initiating more real-world scientific studies.

Back to Top | Article Outline


This work was completed with the support of National Institute of Mental Health grants #1ROI MH49958-04 and P30 MH58107.

Back to Top | Article Outline


1. Branson BM, Handsfield H, Lampe MA, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Morb Mortal Wkly Rep. 2006;55:1-17.
2. Lyons MS, Lindsell CJ, Ledyard HK, et al. Emergency department HIV testing and counseling: an ongoing experience in a low-prevalence area. Ann Emerg Med. 2005;46:22-28.
3. Brown J. Routine HIV screening in the emergency department using the new CDC guidelines: results from a high prevalence area. J Acquir Immune Defic Syndr. 2007;46:394-400.
4. Lyss SB, Branson BM, Kroc KA, et al. Detecting unsuspected HIV infection with a rapid, whole-blood HIV test in an urban emergency department. J Acquir Immune Defic Syndr. 2007;44:435-442.
5. Centers for Disease Control and Prevention. Rapid HIV testing in emergency departments-three U.S. sites, January 2005-March 2006. MMWR Morb Mortal Wkly Rep. 2007;56:597-601.
6. Branson BM. To screen or not to screen: is that really the question? Ann Intern Med. 2006;145:857-859.
7. Gray RH, Makumbi F, Serwadda D, et al. Limitations of rapid HIV-1 tests during screening for trials in Uganda: diagnostic test accuracy study. BMJ. 2007;335:188-191.
8. Rotheram-Borus MJ, Newman P, Etzel M. Effective detection of HIV. J Acquir Immune Defic Syndr. 2000;25(Suppl 2):S105-S114.
9. Rothman R, Lyons M, Haukoos J. Uncovering HIV infection in the emergency department: a broader perspective. Acad Emerg Med. 2007;14:653-657.
© 2007 Lippincott Williams & Wilkins, Inc.