To the Editors:
New York State has the highest number of HIV/AIDS cases in the country.1 In 2006, 1591 AIDS diagnoses in New York occurred within 1 year of HIV diagnosis, accounting for 35% of all new HIV diagnoses.2 In response to this high number of late diagnoses, the advent of rapid and accurate screening tests and a recommendation from the Centers for Disease Control and Prevention to routinely offer HIV testing to adults in all health care settings,3 HIV testing in New York City has expanded to the emergency department (ED). Despite recent efforts to simplify New York State legal mandates for HIV pretest counseling and signed consent,4 the existence of these mandates has prompted some programs to continue to rely on dedicated HIV counselors for testing in the ED. The additional value of these ED testing programs remains controversial. Our evaluation of a counselor-based ED testing program supports the premise that the ED is an important site for HIV testing because we found that patients who did not access care in other medical center settings were tested and successfully referred for follow-up.
The Counseling and Testing Service (CTS) is located at Columbia University Medical Center (CUMC), an academic teaching hospital in northern Manhattan. It was created in 2002 to assist providers with HIV counseling and testing, ensure that all patients who test positive receive their results, and report all newly diagnosed cases of HIV to the New York City Department of Health and Mental Hygiene. Given that the CUMC ED and the ED of an affiliated community hospital 3 miles to the north serve a diverse community of socioeconomic, ethnic, and HIV risk factor groups where late diagnosis of HIV is common, CTS obtained funding from the New York City Department of Health and Mental Hygiene in January 2006 for free rapid HIV testing in these 2 EDs.
The grant funds a counselor for each ED during business hours Monday to Friday. ED staff are encouraged to refer patients for HIV testing, but counselors approach the majority of patients without regard to HIV risk factors, thus patient selection occurs via both nontargeted screening and diagnostic testing.5 Counselors conduct point-of-care testing using the OraQuick ADVANCE (OraSure Technologies, Bethlehem, PA) on fingerstick blood. In the event of a preliminary positive OraQuick result, they draw blood for a confirmatory Western blot. In most cases, patients return to the same CTS counselor for confirmatory results and active referrals5 to care. CTS tracks patients for up to 3 months to ensure that they have kept an appointment with an HIV provider.
For patients counseled and tested from January 1, 2006 through December 31, 2007, data extracted from the CTS program database included demographics, test result, new vs. known HIV diagnosis, transmission risk factors in new HIV diagnoses, and linkage to care, defined as either an electronic medical record (EMR) indication of a visit with an HIV provider at CUMC or patient self-report of a visit outside of the institution within 90 days after receiving confirmatory test results. For patients with more than 1 testing encounter in the specified period, the result of the first test was recorded and subsequent tests were excluded from the analysis, thus the patient was primary unit of analysis (Table 1). An EMR review was performed for patients with a preliminary positive Oraquick result to confirm their HIV status, obtain CD4 cell count and HIV RNA level within 3 months of diagnosis, and record encounters at CUMC in the year before diagnosis. To compare patients newly diagnosed in the ED with patients diagnosed elsewhere in the institution, the EMR was also used to obtain the age, gender, race, and CD4 count, if available, on all new diagnoses of HIV at CUMC in 2006-2007. Data were analyzed using Stata/SE 10. This study was approved by the CUMC Institutional Review Board.
Of 69,398 ED visits during the hours of testing program operation, 2569 patients underwent rapid HIV testing (3.7%). Preliminary positive results were obtained on 31 patients [1.2%, 95% CI (confidence interval): 0.8% to 1.7%]; all diagnoses were confirmed by Western blot, in contrast to recent reports of false positive test results in the ED with oral fluid Oraquick.6 Seven of these patients were found by EMR review to have been previously diagnosed as HIV infected, and 24 (0.9%, 95% CI: 0.6% to 1.4%) were considered new diagnoses. Although data on refusal of offers of HIV testing were not kept for the entire period, a quality assurance project conducted in December 2007 found that of 368 patients offered rapid HIV testing by the counselors, 104 (28.3%) accepted. The most common reasons for refusing testing were that patients did not view themselves at risk or reported a prior test, consistent with data from other ED testing programs.7
Of the 24 newly diagnosed patients, 16 (67%) were black or Hispanic heterosexuals without a history of intravenous drug use. Ten patients (42%) had a CUMC visit in the year before diagnosis; 9 of these patients had at least 1 previous ED visit, and 7 had previously visited only the ED (mean number of visits = 1.7). Ten patients (42%) were admitted to an inpatient medicine service from the ED at the time of HIV diagnosis; all of these patients except 1, who preferred to follow up out of state, were documented as having been linked to care after discharge. Another 8 patients who were not admitted were linked to care, for a total of 17 patients (71%). Within 3 months of diagnosis, CD4 cell count data were available for 17 of 24 patients (median = 200 cells per microliter, range 4-516) and HIV RNA data for 16 patients, 4 of whom had an HIV RNA level greater than 100,000.
CTS staff performed extensive outreach to locate the 7 patients whose follow-up outcomes are unknown. One patient made several CUMC HIV clinic appointments but did not keep them, 1 reported planned follow-up with a private physician, 1 kept a social work appointment only, 1 told CTS he was not ready to be in care, 1 hung up the phone on the CTS counselor, and 1 moved to another state at the time of discharge from the hospital. There was 1 patient for whom no further information was available.
In our analysis of 2569 ED patients, there were 41 repeat testers (1.6% 95% CI: 1.1% to 2.2%), one of whom tested positive on a repeat ED visit, for a total of 25 new ED HIV diagnoses. Thus, ED HIV diagnoses accounted for 16.5% of the 152 new HIV cases diagnosed by conventional enzyme immunoassay or rapid test at the medical center in 2006-2007, whereas ED HIV tests accounted for 5.5% of total medical center tests (conventional enzyme immunoassay and rapid tests). Acknowledging that some proportion of tests from other medical center sites are ordered to confirm infection or also represent repeat testers, the yield of testing patients for HIV in the ED seems to be substantial compared with the institution as a whole.
To see if there were differences between newly diagnosed patients by institutional testing site, we compared the age, gender, race, and CD4 cell count of patients newly diagnosed with HIV in the ED (n = 25) with those diagnosed in outpatient (n = 66) and inpatient (n = 61) settings. The mean age of inpatients was slightly older (43.2 vs. 37.2 years, P < 0.001), more women were diagnosed as outpatients than in the ED (45.5% vs. 20%, P = 0.03), and there were no significant differences by race. CD4 cell count data within 3 months of diagnosis was available for 90.2% patients diagnosed as inpatients and 72.7% of patients diagnosed as outpatients. Mean CD4 cell count of patients diagnosed in the ED (216 cells/μL) was similar to that of inpatients (177 cells/μL) but lower than that of those diagnosed as outpatients (428 cells/μL, P < 0.001).
Our analysis suggests that a counselor-based ED rapid HIV testing program can contribute substantially to HIV screening efforts, consistent with other literature.8 The ED may identify persons with previously undiagnosed HIV infection who do not access care in other medical center settings and who may have been missed by a traditional risk factor-based approach to testing. In addition, CD4 counts at the time of diagnosis may be lower than those of outpatients. This work also adds to the ED testing literature by providing detailed linkage to care outcomes and demonstrating that 100% linkage to HIV care may be difficult, despite extensive testing program outreach to newly positive patients. Additional research is needed on interventions that can link and retain newly positive patients in care.
We thank CTS counselors Paula Merricks-Lewis, Gilbert Serrano, Margarita Delarosa, and Gladys Ojeda, and Ramon Negron, for their assistance with data management.
Katerina A. Christopoulos, MD, MPH*
Bruce R. Schackman, PhD†
Gunjeong Lee, PhD‡
Robert A. Green, MD, MPH§
Ellen A.B. Morrison, MD, MPH∥
*HIV/AIDS Division, San Francisco General Hospital, University of California San Francisco, San Francisco, CA
†Department of Public Health, Weill Cornell Medical College, New York, NY
‡Department of Sociomedical Sciences, Columbia University Mailman School of Public Health, New York, NY
§Division of Emergency Medicine, New York-Presbyterian Hospital/Columbia University Medical Center, New York, NY
∥Division of Infectious Diseases, Department of Medicine, Columbia University College of Physicians and Surgeons, New York, NY
3. Branson BM, Handsfield HH, Lampe MA, et al. Centers for Disease Control and Prevention (CDC). Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep
. 2006;55 (RR-14):1-17.
4. Testimony of Thomas R. Frieden, MD, MPH, Commissioner, New York City Department of Health and Mental Hygiene, before the New York State Assembly Committee on Health on HIV Counseling, Testing, and Informed Consent, December 20, 2006. New York City Department of Health and Mental Hygiene, Bureau of Intergovernmental Affairs. Available at: http://www.nyc.gov/html/doh/html/testi/testi20061220.shtml
. Accessed September 2, 2009.
5. Lyons MS, Lindsell CJ, Haukoos JS, et al. Nomenclature and definitions for emergency department human immunodeficiency (HIV) testing: report from the 2007 conference of the national emergency department HIV testing consortium. Acad Emerg Med
6. Walensky RP, Arbelaez C, Reichmann WM, et al. Revising expectations from rapid HIV tests in the emergency department. Ann Intern Med
7. Brown J, Kuo I, Bellows J, et al. Patient perceptions and acceptance of routine emergency department HIV testing. Public Health Rep
8. Walensky RP, Arbelaez C, Reichmann W, et al. Counselor vs. Provider Based HIV Testing in the Emergency Department: Results from the Universal Screening for HIV in the Emergency Room Trial [abstract]. Presented at: 16th Conference on Retroviruses and Opportunistic Infections; February 8-11,2009; Montreal, Canada.