Skip Navigation
Service Delivery Innovation Profile

Emergency Department Uses Tool To Identify At-Risk Patients in Need of HIV Testing, Leading to Same Number of Newly Diagnosed Patients with Fewer Screening Tests


Tab for The Profile
Comments
(0)
   

Snapshot

Summary

Instead of attempting to screen all patients for HIV, the Denver Health Medical Center emergency department uses an electronic risk assessment tool known as the Denver HIV Risk Score to estimate a patient's risk and hence determine who should be tested. The tool covers three demographic and five behavioral risk factors, each of which is assigned a point value, with the cumulative score reflecting the patient’s overall estimated level of risk. A triage nurse administers the tool during the patient intake process and documents responses in an electronic tracking system that calculates the score in real time. The nurse offers rapid HIV testing to all patients whose score indicates a moderate or higher risk of HIV. Consenting patients have their blood drawn in the treatment area by a nurse, with results returned by the hospital laboratory within 40 minutes and given to the patient while still in the ED. Compared with nontargeted screening, the program identified the same number of HIV-positive individuals, even though it screened significantly fewer patients. These findings suggest that the program may be more effective than nontargeted screening.

Evidence Rating (What is this?)

Moderate: The evidence consists of a comparison of the number and proportion of rapid HIV tests that turn out to be positive under two different approaches to screening in the emergency department: targeted screening of at-risk patients based on the results of an electronic risk assessment tool and nontargeted screening of all patients.
begin do

Developing Organizations

Denver Health
end do

Use By Other Organizations

Several institutions, including the University of Arizona Medical Center in Tucson, are building the risk assessment tool into their electronic medical records.

Date First Implemented

2010
January

Problem Addressed

Many individuals have HIV but do not know it, leading to negative health consequences for themselves and potentially others. Recommendations to screen most patients in most health care settings as a routine part of care are often not followed, due in large part to cost and efficiency concerns. Targeting high-risk individuals for screening may be more practical and less costly, but validated criteria for determining who should be screened do not yet exist. 
  • Many undiagnosed: Approximately 1.1 million adults and adolescents in the United States are HIV-positive, but approximately 20 percent do not know they have the infection.1 As a result, they do not get treatment that could benefit their health or take action to prevent the spread of HIV to others.
  • Lack of adherence to nontargeted screening recommendation: In 2006, the Centers for Disease Control and Prevention (CDC) recommended routine screening of most patients in most health care settings, with patients given the option to opt out of the test if they so choose.2 However, many provider organizations do not adhere to this recommendation, due in part to concerns about the high costs and inherent inefficiencies associated with nontargeted screening, which requires that thousands of low-risk individuals get screened in order to identify a relatively small number of previously undiagnosed cases.3,4
  • Targeted testing possibly more cost-effective: An analysis of CDC recommendations found that, compared with nontargeted testing, targeted testing services would result in more new HIV diagnoses (188,170 versus 56,940) and prevent more HIV infections (14,553 versus 3,644) at a lower cost per infection averted ($59,383 versus $237,149).5
  • Lack of criteria to allow for targeted screening: Targeted screening of high-risk individuals could potentially be a more practical, less costly approach than nontargeted screening, but validated criteria for identifying patients in need of screening did not exist prior to implementation of this program.3,4

What They Did

Back to Top

Description of the Innovative Activity

The Denver Health Medical Center emergency department (ED) uses an electronic assessment tool to estimate a patient's risk of HIV and determine who should be tested. The tool covers three demographic and five behavioral risk factors, each of which is assigned a point value, with the cumulative score reflecting the patient’s estimated level of risk. A triage nurse administers the tool during the intake process and documents responses in an electronic tracking system that calculates the score in real time. The nurse offers rapid HIV testing to all patients whose score indicates a moderate or higher risk of HIV. Consenting patients have their blood drawn in the treatment area, with results returned within 40 minutes and given to the patient while still in the ED. Key program elements include the following:
  • Administration of electronic tool during intake process: The risk assessment tool has been integrated into the ED’s electronic patient tracking system. During the intake process, the system prompts the ED triage nurse to administer the tool by asking the patient a series of eight questions, each covering a separate risk factor for HIV. Nurses can choose not to use the tool in certain circumstances, such as a patient expressing privacy concerns or being too acutely ill and/or in too much pain to answer the questions. Three questions cover demographic factors (age, gender, and race/ethnicity) and five cover health-related behaviors, such as whether the individual has been tested for HIV in the past, used injectable drugs, or engaged in risky sexual behaviors, such as men having had sex with other men and anal intercourse. Each question has an assigned point value, yielding a cumulative score of –14 to +81. Two risk factors (male gender and men having sex with other men) are considered the biggest risks and hence have been assigned the highest number of points.3
  • Automated, real-time scoring to identify at-risk individuals: After each question, the nurse inputs the patient’s response into the electronic system, which automatically calculates the risk score in real time. The system places patients into one of five risk categories based on the cumulative score: under 20 (indicating very low risk), 20 to 29 (low risk), 30 to 39 (moderate risk), 40 to 49 (high risk), and 50 and above (very high risk). Anyone scoring 30 or above becomes a candidate for screening. In practice, the nurse will stop asking questions once a patient reaches 30 points or when it becomes impossible for a patient to reach that level, since in both instances asking the remaining questions will not change the ultimate decision as to whether to proceed with HIV testing. As a result, the last one or two questions are not asked approximately 15 to 20 percent of the time.
  • Offer of screening to at-risk individuals: For any patient with a cumulative score of 30 points or higher, the system prompts the triage nurse to ask the patient if he or she would like a rapid HIV blood test. If the patient agrees, the triage nurse checks the “yes” box on the screen, which triggers a test order for the ED nurse.
  • Test administration and results during ED visit: For those who consent to testing, the ED nurse draws blood in the treatment area and sends the sample to the hospital’s laboratory. The laboratory completes the test and returns the results within 40 minutes, with the ED nurse then delivering the results to the patient.
  • Counseling and followup for those testing positive: Patients identified with HIV infection are provided standardized post-test counseling by ED-based clinical social workers who also link them into ongoing medical care. 

Context of the Innovation

An integrated system that serves as Colorado's largest public safety net institution, Denver Health operates the Denver Health Medical Center, a 477-bed urban public hospital that has approximately 110,000 ED and urgent care patient visits each year. Many of these patients have risk factors for HIV. Also part of Denver Health, Denver Public Health is responsible for developing and implementing public health initiatives in conjunction with the Denver County Department of Public Health and in collaboration with the State public health department.

The impetus for this program came from Denver Health ED clinicians, who were evaluating implementation of the aforementioned 2006 CDC recommendations.6 Their analysis concluded that the CDC's recommended approach was impractical from both time and cost perspectives. This determination, along with the existence of a detailed HIV testing registry at Denver Health, convinced the clinicians of the need to develop a more targeted, data-driven approach to HIV testing.

Did It Work?

Back to Top

Results

Compared with CDC-recommended nontargeted screening, the program identified the same number of HIV-positive individuals, even though it screened significantly fewer patients. These findings suggest that the program may be more effective than nontargeted screening.
  • Equally effective in identifying HIV-positive individuals: In a comparison of nontargeted screening and targeted screening with the assessment tool (used during separate 4-month periods in the ED), both approaches identified the same number of newly diagnosed HIV cases (seven).
  • Significantly fewer screening tests: While using the targeted approach, the ED identified and screened 551 moderate- to high-risk patients, less than one-sixth the number screened when the nontargeted approach was employed (3,591 patients).
  • Higher proportion of HIV-positive patients, suggesting greater cost-effectiveness: With the targeted program, 1.3 percent of all screened patients tested positive for HIV, more than six times the 0.2 percent of patients testing positive with nontargeted screening. Although the researchers did not conduct a formal analysis, these data suggest that targeted screening may be more cost effective, since it can identify the same number of new HIV cases with significantly fewer tests. Consequently, the program has the potential to reduce costs.

Evidence Rating (What is this?)

Moderate: The evidence consists of a comparison of the number and proportion of rapid HIV tests that turn out to be positive under two different approaches to screening in the emergency department: targeted screening of at-risk patients based on the results of an electronic risk assessment tool and nontargeted screening of all patients.

How They Did It

Back to Top

Planning and Development Process

Selected steps included the following:
  • Using existing registry: Investigators pulled data from the existing Denver Health HIV registry, which included data from approximately 92,635 patients age 13 and older who had come to Denver Public Health for outpatient HIV testing between 1996 and 2008. The database includes the following information, collected from each patient prior to testing: demographic characteristics, symptoms, history of sexually transmitted infections, sexual history, specific sexual practices, gender of sexual contacts, previous HIV testing history, and other risk factors associated with HIV transmission (e.g., use of injectable drugs, prostitution, sexual contact with high-risk individuals).
  • Developing and validating tool: Investigators used multivariable logistic regression analysis to develop the tool, analyzing more than 50 possible variables. The goal was to develop a very simple tool (i.e., one that uses the fewest possible variables) with strong predictive ability. As discussed earlier, the investigators ultimately identified eight variables that collectively yield high predictive validity for HIV risk. They validated use of the tool in a test involving 22,983 patients at the University of Cincinnati Medical Center ED, which put a targeted HIV screening program in place in 1998.
  • Implementing tool in existing systems and processes: Information technology (IT) staff integrated the tool into the Denver Health Medical Center ED’s electronic tracking system, and tool developers worked with ED-based nurses to incorporate the screening questions and the blood draw into existing work processes.
  • Training triage nurses: Triage nurses were provided with informal training from program developers on how to use the tool during medical screening/triage.
  • Evaluating and refining tool: As noted, predictive validity was evaluated in the previously described study that compared use of the tool with use of nontargeted screening. Since completion of that study, additional analysis suggests that the last two questions (which cover history of vaginal and anal intercourse) can be eliminated without diminishing the tool's predictive validity. Investigators plan to remove these questions from the tool and re-evaluate its effectiveness; results will likely be published in 2017.

Resources Used and Skills Needed

  • Staffing: The program requires no new staff, because existing staff integrate the administration of the tool and HIV testing into their regular responsibilities.
  • Costs: The tool cost roughly $150,000 to $250,000 to develop and evaluate. (More information can be found in the Funding Sources section below.) Organizations interested in using the tool can obtain it free of charge. Adoption costs tend to be quite small, consisting primarily of integrating the tool into an existing electronic patient tracking or medical record system. Ongoing use of the tool entails no significant costs.
begin fsxml

Funding Sources

Agency for Healthcare Research and Quality; National Institute of Allergy and Infectious Diseases; Colorado Department of Public Health and Environment
The details of the support from these organizations for the development and validation of the Denver HIV Risk Score are presented below: 
  • The Agency for Healthcare Research and Quality provided a 5-year grant totaling $515,028 (K02 HS017526).
  • The Colorado Department of Public Health and Environment provided a $358,187 grant.
  • The National Institute of Allergy and Infectious Diseases provided a 5-year grant totaling $3,157,714 (R01 AI106057).
end fs

Tools and Other Resources

The Denver HIV Risk Score can be found in the following publication:
Haukoos JS, Hopkins E, Bender B, et al. Comparison of enhanced targeted rapid HIV screening using the Denver HIV risk score to nontargeted rapid HIV screening in the emergency department. Ann Emerg Med. 2013;61(3):353-61. [PubMed]

Adoption Considerations

Back to Top

Getting Started with This Innovation

  • Obtain support of hospital and ED leaders: Planners should seek the support of hospital and ED leaders by highlighting the benefits of adopting more targeted screening, including the results from Denver Health Medical Center's experience, which suggest the potential for significant cost savings with no negative impact on care processes or patient throughput.
  • Explain value of tool to nurses: Adopters should explain the effectiveness of targeted screening to nurses when teaching them to use the tool.
  • Work with IT staff to incorporate tool into electronic systems: IT staff must be enlisted to incorporate the tool into existing electronic systems. Since the tool is not complicated, this process tends to be fairly straightforward.
  • Build flexibility into system: Nurses should be permitted to decide not to use the tool in circumstances where its use is not practical or advisable, such as when the patient expresses concerns about privacy or is too acutely ill or in too much pain to answer the questions.

Sustaining This Innovation

  • Embed tool into existing work processes: Clinicians and other staff will be much more likely to continue using the tool if it is integrated into existing clinical care and work processes.
  • Monitor and share information with staff: Monitoring information about the program's impact and sharing this information with staff members will encourage them to sustain its use.
  • Refine tool based on new information: Clinicians should keep abreast of current HIV research and refine the tool accordingly.

Use By Other Organizations

Several institutions, including the University of Arizona Medical Center in Tucson, are building the risk assessment tool into their electronic medical records.

More Information

Back to Top

Contact the Innovator

Jason S. Haukoos, MD, MSc
Department of Emergency Medicine
Denver Health Medical Center
777 Bannock Street, Mail Code 0108
Denver, CO 80204
(303) 602-5174
E-mail: jason.haukoos@dhha.org

Innovator Disclosures

Dr. Haukoos reported having no financial interests or business/professional affiliations relevant to the work described in the profile, other than the funders listed in the Funding Sources section.

References/Related Articles

Haukoos JS, Hopkins E, Bender B, et al. Comparison of enhanced targeted rapid HIV screening using the Denver HIV risk score to nontargeted rapid HIV screening in the emergency department. Ann Emerg Med. 2013;61(3):353-61. [PubMed]

Haukoos JS, Lyons MS, Lindsell CJ, et al. Derivation and validation of the Denver Human Immunodeficiency Virus (HIV) risk score for targeted HIV screening. Am J Epidemiol. 2012;175(8):838-46. [PubMed]

Haukoos JS, Hopkins E. Understanding HIV screening in the emergency department: is perception reality? Acad Emerg Med. 2013;20(3):309-12. [PubMed]

Haukoos JS, White DA, Lyons MS, et al. Operational methods of HIV testing in emergency departments: a systematic review. Ann Emerg Med. 2011;58(1Suppl1):S96-103. [PubMed]

Footnotes

1 Chen M, Rhodes P, Hall HI, et al. Prevalence of undiagnosed HIV infection among persons aged ≥13 Years—National HIV Surveillance System, United States, 2005–2008. MMWR Morb Mortal Wkly Rep. 2012;61Suppl:57-64. [PubMed]
2 Branson BM, Handsfield HH, Lampe MA, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep. 2006;55(RR-14):1-17. [PubMed]
3 Haukoos JS, Hopkins E, Bender B, et al. Comparison of enhanced targeted rapid HIV screening using the Denver HIV risk score to nontargeted rapid HIV screening in the emergency department. Ann Emerg Med. 2013;61(3):353-61. [PubMed]
4 Haukoos JS, Lyons MS, Lindsell CJ, et al. Derivation and validation of the Denver Human Immunodeficiency Virus (HIV) risk score for targeted HIV screening. Am J Epidemiol. 2012;175(8):838-46. [PubMed]
5 Holtgrave DR. Costs and Consequences of the US Centers for Disease Control and Prevention's Recommendations for Opt-Out HIV Testing. PLOS Med. 2007;4(6):e194. [PubMed]
6 Haukoos JS, Hopkins E, Conroy AA, et al. Routine opt-out rapid HIV screening and detection of HIV infection in emergency department patients. JAMA. 2010;304(3):284-92. [PubMed]
Comment on this Innovation

Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.

Original publication: April 23, 2014.
Original publication indicates the date the profile was first posted to the Innovations Exchange.

Last updated: April 23, 2014.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.