Special Focus: HIV Screening
From the March ACP Observer, copyright © 2007 by the American College of Physicians.
In September 2006, the Centers for Disease Control and Prevention (CDC) issued new guidelines calling for universal HIV testing of all Americans between the ages of 13 and 64. It was an aggressive step by the CDC to promote earlier diagnosis of HIV, thus potentially slowing transmission by those who don't know they are infected, and getting infected patients into treatment before they develop full-blown AIDS.
According to an editorial published in the December 5, 2006, Annals of Internal Medicine, each year an estimated 16 million to 18 million adults are tested for HIV infection. As of 2002, 68 million people reported that they had been tested at least once. That is 38% of U.S. adults age 18 to 64.
Testing has successfully identified HIV in approximately 75% of infected patients, according to the editorial. However, nearly 40% of infected patients carry the virus for a decade before they develop symptoms and are tested for HIV. In the interim, many likely have already transmitted the infection unwittingly.
The new CDC guidelines suggest that physicians eliminate signed consent and proceed with testing unless a patient actively opts out. The agency also suggests that physicians dispense with pre-test counseling, except in high-risk settings such as STD clinics.
"People know about HIV and what the risks are," said Roger Chou, MD, a principal investigator at the Oregon Evidence-Based Practice Center at Oregon Health and Science University in Portland, Ore., lead investigator for the evidence review used to develop HIV screening recommendations by the U.S. Preventive Services Task Force, and a consultant for ACP's PIER HIV screening module. "But we're still not making a dent in the new infections. It's been about 40,000 per year for the last 10 years or so."
Many infected persons remain undiagnosed for years even though they have multiple encounters with the health care system, according to the CDC. Between 1990 and 1992, the proportion of persons who first tested positive for HIV less than a year before receiving a diagnosis of AIDS was 51%. That proportion declined only to 39% by 2004.
According to a report published in June 2003 in the CDC's Morbidity and Mortality Weekly Report, people who had their first positive HIV test less than a year before receiving a diagnosis of AIDS were:
- African-American (56%) or Hispanic (23%)
- exposed through heterosexual contact (44%)
- tested for HIV antibody because of illness (65%)
- first found to be HIV positive at an acute or referral medical care facility (87%)
With the advent of effective treatment, clinicians and public health officials have increasingly called for expanded routine screening. Dr. Chou noted that the new CDC recommendations could lessen the stigma of HIV screening by expanding routine screening beyond high-risk settings.
"The hope is that we find more people with HIV, counsel them and get them into treatment, so they can stop transmitting infection," Dr. Chou said. "Eventually, that would decrease the overall health care burden by decreasing the incidence of HIV."
However, the new CDC recommendations for HIV screening do not take into account the patchwork of legal requirements that still exist in nearly all states that still mandate various forms of pretest counseling and formal consent processes. The process of reconciling the new guidelines with state laws will take time, and physicians are advised to follow the situation closely. (See sidebar).
This edition of ACP Observer Special Focus is designed to optimize your ability to screen patients for HIV.
WHOM TO SCREEN
The newly revised CDC recommendations advocate voluntary HIV screening become a routine part of medical practice, similar to screening for other treatable conditions. These latest recommendations for testing address all health care providers in the public and private sectors, including those working in hospital emergency departments, urgent care clinics, inpatient services, substance abuse treatment clinics, public health clinics, community clinics, correctional health care facilities and primary care settings.
CDC's HIV screening recommendations: What's changed?
Screening in all health care settings, including physician practices, for all individuals ages 13 to 64, regardless of perceived or reported risk.
Previous: Routine HIV testing only of individuals with recognized risk factors or in settings where the disease prevalence exceeds 1%. (See sidebar "What increases the risk for HIV infection".)
Voluntary opt-out approach to testing to "normalize" screening and testing as part of routine care.
Previous: Separate informed consent specific to HIV testing and pre- and post-test counseling.
Expanded testing and screening for pregnant women. Rapid testing during labor for women of unknown HIV status. Repeat HIV screening in the third trimester for women with known risk factors.
Previous: Opt-out approach to testing for pregnant women.
HOW TO SCREEN
Consent and pretest information
The new CDC guidelines propose that patients should be informed orally or in writing that HIV testing will be performed as part of routine screening unless they decline. Oral or written information should include an explanation about HIV infection and the implications of positive and negative test results, and the patient should be offered an opportunity to ask questions and to decline testing. With such notification, consent for HIV screening should be considered incorporated into the patient's general informed consent for medical care on the same basis as other screening or diagnostic tests; a separate consent form for HIV testing is not required.
Easily understood informational materials should be made available in the languages of commonly encountered populations within the service area. Availability of interpreters and bilingual staff to provide language assistance to patients with limited English proficiency should be ensured. If a patient declines an HIV test, the decision should be documented in the medical record, as for any other test the patient declines.
Standard blood testing algorithm
Use a third-generation HIV-1 enzyme immunoassay (EIA) antibody test to screen for HIV infection. A negative HIV antibody test indicates that the patient is unlikely to be infected; however, there is a window of 22 days between time of infection and the detection of antibodies during which the antibody test may be negative. The sensitivity of third-generation HIV-1 antibody tests is 99.4% to 100%, and specificity is 99.5%; sensitivity is lower in early HIV infection.
If the initial HIV test result is positive, then the Western blot or, less commonly, the immunofluorescent antibody (IFA), is the gold standard for confirmation of infection, with a specificity of 99.95%. Currently, all enzyme immunoassay (EIA) HIV-positive results are confirmed by a positive Western blot or IFA.
A Western blot (shown, 48 specimens and 2 controls). Western blot testing sets the gold standard for confirmation of HIV infection, with a specificity of 99.95%. Currently, all HIV-positive enzyme immunoassay (EIA) results are confirmed by a positive Western blot or immunofluorescent antibody (IFA) test.
Recognize that the patient is unlikely to be infected if the Western blot is negative, or persistently indeterminate in a patient with no risk factors for HIV infection, even if the initial EIA was positive.
Do not use p24 antigen, HIV RNA PCR or HIV DNA PCR testing to screen asymptomatic children or adults for HIV disease. (HIV RNA PCR or HIV DRNA PCR is used to screen neonates of mothers with known HIV infection or unknown serostatus at the time of delivery, and p24 antigen is used as a diagnostic test in patients with suspected very early primary HIV infection.)
A complete list of FDA-approved HIV-1 and HIV-1/2 tests can be found online.
Rapid EIA tests are performed on serum, plasma or whole blood. A presumptive diagnosis can be made within 30 minutes, which can be very useful in clinics where patients might not return for results. Rapid testing can also be used after an occupational exposure to HIV infection or when the labor and delivery setting provides the first opportunity to test.
Estimates of sensitivity and specificity of rapid HIV tests (prior to confirmatory testing) are similar to those for standard antibody testing, although false positives can occur, particularly in lower prevalence settings. In such settings, the negative predictive value of a single rapid test is high.
Confirm all results using standard testing algorithms. However, positive results from rapid HIV tests can be released to the patient before confirmatory testing if the patient will benefit from the result and appropriate counseling is provided.
The FDA has approved the following rapid immunoassay tests: Murex SUDS HIV-1 Test, Reveal Rapid HIV-1 Antibody Test, Uni-Gold Recombigen HIV, Multispot HIV-1/HIV-2 Rapid Test, OraQuick ADVANCE Rapid HIV-1/2 Antibody Test, and HIV 1/2 STAT-PAK ASSAY and SURE CHECK HIV 1/2 ASSAY. Two of these tests, Uni-Gold and OraQuick, have been granted a Clinical Laboratory Improvement Amendments waiver from the FDA. This designation allows them to be used for true point-of-care testing at the bedside, with results available in approximately 10 to 30 minutes.
Saliva and urine testing
HIV screening can be performed on urine or saliva rather than blood and may improve testing uptake in some patient groups, particularly adolescents and young adults. In one study, health educators, using a confidential survey, showed that young people between the ages of 12 and 24 preferred any of three salivary tests to one using urine or two others requiring fingersticks, especially when results could be obtained rapidly.
The sensitivity of salivary sampling for HIV testing is 96% to 100%, the specificity is >99.9%, and it is a method easily accepted by patients; however, there are more indeterminate Western blot results. The sensitivity and specificity using urine samples are 98% to 99%. Repeated testing of positive samples using serum is advised due to the lower specificity of Western blot in urine. Thus, because of lower accuracy, urine HIV testing is not widespread in the U.S.
Tests that facilitate collection of dried blood samples at home and are then sent for laboratory testing are 100% sensitive and specific compared with venous blood samples and may provide patients with an added measure of privacy. One such test, the Home Access Express HIV-1 Test system, is approved for use in the U.S. Home tests that do not include laboratory testing should not be utilized for screening because they do not produce reliable results and are not approved by the FDA.
According to PIER, the CDC HIV Counseling, Testing, and Referral Services recommends retesting individuals with ongoing exposure to HIV. Frequency of testing depends on several factors, including the type and timing of exposure, client preferences and local resources for testing. In general, perform repeat testing at 3 and 6 months after any recent exposure to HIV in patients with a negative initial HIV antibody test result and in those who are HIV negative but continue to practice high-risk behaviors or are in a population with a high prevalence (>1%) of HIV. (See sidebar "What increases the risk of HIV infection"). In addition, health care providers should repeat testing at 1 month in persons who are HIV antibody positive and indeterminate by Western blot.
All persons likely to be at high risk for HIV should be retested at least annually, according to the CDC's new guidelines. This includes injection-drug users and their sex partners; persons who exchange sex for money or drugs; sex partners of HIV-infected persons; and MSM or heterosexual persons who themselves or whose sex partners have had more than one sex partner since their most recent HIV test. Health care providers should encourage patients and their prospective sex partners to be tested before initiating a new sexual relationship.
For patients with no known exposure and who are not in a population with a high prevalence of HIV infection, repeated testing after a negative initial HIV antibody test is not required.
Routinely offer HIV screening to all pregnant women, regardless of the presence or absence of risk factors for HIV infection. A discussion of HIV and the benefits of the test increases the rate of consent for testing in pregnant women. Opt-out screening has proven substantially more effective than risk-based testing for detecting unsuspected maternal HIV infection and preventing perinatal transmission.
Screen pregnant women in the first trimester, preferably at the first prenatal visit, and retest pregnant women who continue to practice high-risk behaviors during the third trimester. Repeat screening in the third trimester is recommended in certain jurisdictions with elevated rates of HIV infection among pregnant women. Consider use of commercially available rapid HIV screening tests on pregnant women of unknown serostatus arriving to the delivery ward in active labor.
Screen all neonates of seropositive women or women whose HIV serostatus is unknown. Screening tests and algorithms are different for neonates and are generally managed by neonatologists.
IMPACT OF POSITIVE RESULTS AND PATIENT EDUCATION
There is potential for physical and psychological harm attributable to the diagnosis of HIV infection, especially in women. Counseling or referral to services dealing with partner notification should be performed routinely, according to CDC guidelines. Health departments can assist patients by notifying, counseling and providing HIV testing for partners without disclosing the patient's identity. Providers should inform patients who receive a new diagnosis of HIV infection that they might be contacted by health department staff for a voluntary interview to discuss notification of their partners. If the patient is at risk for domestic violence, prompt referral to appropriate HIV support services should be made. Follow-up for patients who test positive for HIV antibody is crucial. In one study, there was significant delay between learning about a positive test and initiating care, particularly among patients in ethnic minority groups and those without medical insurance.
In studies done before the CDC issued its new guidelines in September 2006, HIV-seropositive women reported greater discrimination in the health care system related to their HIV status (35% vs. 0%), breaches in confidentiality of their HIV-seropositive status to family and friends, and higher psychological stress as measured by the State-Trait Anxiety Inventory (P<0.05). Women reported twice as many violent incidents as men and nearly half reported HIV-positive status as a cause.
One study of HIV-infected women found that 4% reported violence after disclosure of status, and 45% reported emotional, physical or sexual abuse at some time after their diagnosis.
On the other hand, knowledge of HIV status may help reduce high-risk behaviors that are related to transmission. Studies on the effects of counseling and testing for HIV on behavior changes found mixed results, which varied according to the population studied. The most consistent evidence of beneficial effects of counseling and testing came from studies of heterosexual HIV-serodiscordant couples that consistently reported increased condom use. In addition, tested persons who were HIV positive were more likely to adopt less risky behaviors than those who were HIV negative or untested.
Studies have been inconsistent with regard to condom or other birth control method use in HIV-infected women. In intravenous drug users, most studies found reductions in reported drug-related and high-risk sexual practices after HIV counseling and testing. In MSM, there was no consistent evidence on the effects of HIV counseling and testing.
Education regarding the risks of transmission and the medical and social benefits of early HIV diagnosis can increase acceptance of screening, particularly when it is disseminated in a culturally relevant manner, such as using a presenter of the same ethnicity.
This information comes from the PIER module "HIV Screening."
The information included herein should never be used as a substitute for clinical judgment and does not represent an official position of ACP.
Although the CDC guidelines are explicit about whom and when to screen, it is still unclear how physicians will implement them, said Roger Chou, MD, a principal investigator at the Oregon Evidence-Based Practice Center at Oregon Health and Science University in Portland, Ore., and a consultant for ACP's PIER HIV screening module.
For example, the CDC recommendations state that extensive pretest counseling or specific consent for HIV testing is no longer needed. But each institution has to follow state laws that might require otherwise.
"In some states, doing that kind of streamlined testing isn't even legal," Dr. Chou said. "That needs to go through state legislatures."
In the meantime, physicians can use extensive online resources offered by the Health Research and Educational Trust, affiliated with the American Hospital Association, to learn about the laws in different states. Cross-state comparison tables for specific issues, such as counseling requirements, disclosure and confidentiality, and testing pregnant women and newborns, are also available.
Also at issue, Dr. Chou said, is that guidelines issued by the CDC differ in some aspects from those issued by the U.S. Preventive Services Task Force (USPSTF). The USPSTF guidelines are available online.
The USPSTF guidelines don't recommend for or against screening in low-risk persons in low-risk settings because the benefits of screening in these individuals appear small, there are some potential harms, and public health benefits such as decreased transmission have not been clearly demonstrated. So the USPSTF guidelines leave more discretion to clinicians.
"The recommendations differ largely because the standards of evidence used by the CDC and the USPSTF were different," Dr. Chou said. The CDC bases many of its recommendations on potential or likely public health benefits. Many of these assumptions appear quite reasonable at first glance. But upon further consideration, the assumptions are either difficult to prove or studies aren't available yet. He added that the USPSTF in general requires proof of benefits before it will make a recommendation. The CDC is more willing to use modeling or cost-effectiveness data and to extrapolate evidence from other populations, such as prenatal screening or high-risk populations, to low-risk populations and then to apply it more broadly.
"Otherwise the guidelines are quite similar in recommending screening all pregnant women, persons reporting risky behaviors, persons in high-prevalence settings, and persons engaging in high-risk behaviors," Dr. Chou said.
Patient populations in which the risk of HIV infection is known to be high (prevalence > 1%):
- patients with chronic hepatitis B or C,
- psychiatric inpatients or outpatients,
- inmates in correctional facilities,
- those seeking treatment at STD clinics,
- those who use homeless shelters,
- patients in clinics serving men who have sex with men,
- patients in drug or alcohol prevention and treatment programs,
- patients in acute care hospitals, emergency rooms, and other clinical settings
Risk factors for transmission of HIV infection:
- receipt of blood products between 1977 and 1985,
- intravenous drug use with known or potentially contaminated needles,
- percutaneous or mucosal exposure to HIV-infected body fluids,
- history of STD,
- more than one sexual partner
Internist Archives Quick Links
Reviews of the World's Top Medical Journals—FREE to ACP Members!
ACP JournalWiseSM is mobile optimized with optional email alerts! Get access to reviews from over 120 of the world’s top medical journals alerting you to the highest quality, most clinically relevant new articles based on your preferred areas of specialty. ACP Members register your FREE account now!
New CME Option: Internal Medicine 2014 Recordings
New CME Package
Includes 75 of the most popular sessions in internal medicine and the subspecialties. Stream the sessions, answer brief quizzes and earn CME credit. See details.