Please note: In 2003, the CTF updated its Grades of Recommendations to include an "I Recommendation" for situations where insufficient evidence exists to allow a recommendation to be made.  (Formerly, these situations were captured under a "C Recommendation".)  This change is not retroactive, and all "C Recommendations" made prior to 2003 have not been reevaluated in light of the new "I" recommendation grade.  For a discussion of these recommendation grades, please link to the 2003 article in the Canadian Medical Association Journal here.

Screening for HIV Antibody

Overview
Human immunodeficiency virus (HIV) antibody screening may reduce the incidence of acquired immunodeficiency syndrome (AIDS) through counselling to prevent HIV transmission and may delay the onset of AIDS through treatment of asymptomatic HIV-positive patients.

Burden of Suffering
By October 7, 1991, 5,349 AIDS cases had been reported in Canada, including 63 pediatric cases. Sixty four percent of affected adults were ages 20 to 39 years. Sixty percent had died. Under reporting by up to 20% may result from efforts to maintain patient confidentiality and the complexity of the reporting procedure. HIV seroprevalence is 0.3 to 0.6 per 1,000 among pregnant women and 2-3 times higher among urban women.<1-5>

Behaviours that increase risk of HIV infection include having many sexual partners, receiving anal intercourse, sharing needles during injection drug use and receiving blood or blood products prior to introduction of universal HIV antibody screening by blood banks. Homosexual or bisexual men, prostitutes, injection drug users, people receiving blood products between 1978 and 1985, sexual contacts of people with HIV infection and people from countries with a high HIV prevalence are at high risk. Of all cases reported in Canada, 78% were associated with homosexual or bisexual activity only, 1.4% with injection drug use only, and 3.5% with both. 95% of AIDS cases belonged to an identified risk group.

Maneuver
Deciding who should undergo HIV antibody screening is complex and difficult. There are problems with both the frequency with which physicians ask the necessary questions and the quality of responses. A sexual history was routinely obtained by less than half of physicians recently surveyed. Seropositivity rates vary within identified risk groups (related to factors such as race, income, recent injections, or injections in "shooting galleries" within the injection drug use group). Barriers to obtaining accurate information may include worries about self-incrimination, illiteracy, race or ethnic background, language and socioeconomic status.<6> Risk factors were acknowledged in less than 60% of HIV-positive parturient women;<7-10> targeted screening (e.g. premarital testing or in sexually transmitted disease (STD) clinics) has low sensitivity and specificity. The addition of counselling did not improve the sensitivity of questionnaires in identifying seropositive women.<11> Increased patient acceptance of routine testing and the poor accuracy of targeted screening suggest that routine screening is preferable to targeted screening if an elevated seroprevalence rate is suspected in any population.<6-8,10,12>

However, a complete sexual and drug-use history should be obtained for all patients as part of a clinical assessment. The contents of such a history are described in guidelines from the Canadian Medical Association and other groups.

Once a decision is made to screen, informed consent must be obtained, and counselling must be provided before the test and after receipt of the results – an important component of the testing procedure. Maintaining confidentiality is likely to increase the acceptance of testing, although it has not been studied.

Screening for HIV infection currently involves detection of antibodies. The first step requires one of a number of commercially available enzyme-linked immunosorbent assay (ELISA) kits. These are highly sensitive but their specificity is reduced by cross-reactions with components other than HIV antigens. ELISA is easy to perform and inexpensive. If the results are repeatedly positive, the second step involves a more specific confirmatory test, such as the Western blot, radioimmunoprecipitation or immunofluorescence assay. These confirmatory tests are labour intensive and costly. The combined sensitivity of the two tests approaches 100%. However, the trauma to a single individual falsely identified as seropositive may offset the advantage to those who are truly seropositive. The higher the prevalence, the more true positives are identified for every false positive; acceptable prevalence rates to justify screening have not been defined.

False negative results after combined testing with ELISA and the Western blot technique may occur because of the delay in antibody development after HIV exposure. This period is usually less than 6 months. A false negative result may falsely reassure people in whom antibody has not yet developed. Testing should be repeated after 6 months in seronegative people whose behaviour put them at risk.

In neonates inaccurate results may occur if only the antibody but not the virus is passed to the neonate or because of poor antibody response.<13-15> Thus, viral DNA detection through a polymerase chain reaction or virus isolation is required for diagnosis. False-positives may also result from errors in specimen identification or contamination in the laboratory.

Effectiveness of Prevention and Treatment

Risks of Screening
Implications of HIV seropositivity go beyond the infection and relate to behaviours that increase the risk of infection. Being identified as a homosexual man may subject a person to discrimination and prejudice. Behaviours such as illicit injection drug use and prostitution place the seropositive person at risk for legal prosecution. Discrimination in employment, housing, education, health insurance and access to medical care have also occurred among those identified as HIV positive. Rejection by family and friends reduces needed support. Depression and anger are more common in subjects aware of their serologic status than in those unaware of it.<16>

Whether patients who have been engaging in high-risk behaviour are reassured and encouraged to continue these behaviours after receiving a negative test result is unknown but unlikely. Studies of counselling suggest a reduction in behaviour known to transmit the virus, although this reduction is often greater among HIV-positive than HIV-negative people.<16-20>

Benefits of Screening
In asymptomatic people with a history of risk factors, knowledge of being truly seronegative may provide a great deal of relief and influence future plans regarding marriage, children, employment, insurance and protection of sexual partners.

Drug therapy has also been shown to be beneficial. Zidovudine (AZT) delays the onset of AIDS in people with asymptomatic HIV infection with a CD4 lymphocyte count of less than 0.5 X 10 /L.<21>

Optimism about early AZT therapy must be tempered by several considerations. The drug must be given to 100 patients for a year to delay the onset of opportunistic infection in 4 patients. Data are not available on important outcomes such as survival, quality of life and adverse effects. Studies of HIV isolates in treated patients are showing drug resistance. Few data are available for injection drug users, women who have acquired HIV infection heterosexually or children;<21-23> the generalizability of the results to these populations is questionable. In a study involving symptomatic people there was no difference in survival between those given early AZT therapy and those given late therapy despite a delay in the onset of AIDS.<24> More information is needed on long-term efficacy and toxic effects of AZT. The finding that treatment of asymptomatic HIV infection can delay the development of AIDS suggests that the CD4 count should be monitored until it reaches a level at which AZT treatment should be started.

In a randomized trial,<25> the risk of P. carinii pneumonia was reduced by 70% by aerosol pentamidine prophylaxis in asymptomatic patients with a CD4 count less than 0.2 X 10 /L and in patients with AIDS or advanced symptomatic HIV infection who had not had P. carinii pneumonia. Several trials of primary prophylaxis with other similar agents are under way. Previous trials have demonstrated the efficacy of trimethoprim-sulfamethoxazole in other immunocompromised populations.

Screening for and management of concurrent infectious diseases such as tuberculosis (see Chapter 62), fungal infections or parasitic infections may be altered by knowledge of HIV seropositivity. Most recommendations for primary prophylaxis or treatment of these conditions are based on expert opinion, recognizing that conventional management is less efficacious in HIV-positive patients than in HIV-negative subjects.

Efficacy of Counselling
Counselling on high-risk sexual practices or on the use of condoms is expected to increase desired behaviours and decrease the spread of HIV infection. Cohort studies involving homosexual men, have shown a statistically significant reduction in risk behaviour among those aware of their seropositive status.<16-18,26-28> These studies have been criticized because of the highly selected nature of the participants and because of difficulties in measuring and validating sexual behaviour. Studies of counselling people with hemophilia and injection drug users have shown reduction in risk behaviour but have the same design flaws.<20,29-33> Nonetheless, one can conclude that counselling probably reduces but does not eliminate unsafe sexual practices.

Screening of Pregnant Women
Screening of pregnant women is a special situation because the risk of vertical transmission to the fetus is about 30%. One might expect that women found to be seropositive would opt for an elective abortion. However, two studies involving injection drug users showed that seropositive women had the same rate of pregnancy and therapeutic abortion as seronegative women. Nevertheless, knowledge of seropositivity may result in: 1) avoidance of certain obstetric maneuvers (e.g., amniocentesis, chorionic villus sampling and fetal monitoring) that may increase the risk of HIV transmission; 2) earlier diagnosis of associated illnesses that might otherwise be attributed to pregnancy; 3) avoidance of breast feeding, which may be a vehicle for HIV transmission; and 4) earlier treatment of infected infants.

Recommendations of Others
The U.S. Centers for Disease Control (CDC) recommends screening people in identified risk groups, including those with STDs. In 1985 the CDC also recommended screening and counselling for women in high-risk groups to prevent perinatal HIV transmission. Testing and counselling should also be offered to woman who request testing. Testing was not recommended for women at low risk because of difficulties in interpreting the results. In addition, it was recommended that patients being treated for tuberculosis be counselled and undergo HIV antibody testing because of the implications for management. Trimethoprim-sulfamethoxazole or aerosol pentamidine prophylaxis for P. carinii pneumonia has been recommended for adults with a CD4 count of less than 0.2 X 10 /L and trimethoprim-sulfamethoxazole prophylaxis for children with a CD4 count 20% or more below the normal age-specific count.<34,35>

In 1989, the U.S. Preventive Services Task Force<36> concluded that there was fair evidence to recommend screening in a high-risk population and in pregnant women but no evidence to either include or exclude screening in a low-risk population.

Conclusions and Recommendations
Obtaining a history of sexual behaviour and injection drug use and offering counselling has limited sensitivity for identifying HIV-positive people in the general population, but is likely to increase detection of risk behaviours. Its inclusion in the periodic health examination of asymptomatic people in the general population is based on expert opinion (C Recommendation).

Recommendations for HIV antibody screening must consider characteristics of the screening maneuver, particularly sensitivity and specificity, and the availability of treatment for asymptomatic seropositive people. There is insufficient evidence to recommend the inclusion or exclusion of HIV antibody screening among pregnant women (C Recommendation). Because the prevalence of HIV infection is lower in Canada than in the U.S. the generalizability of the results of U.S. studies is questionable. Even with excellent test characteristics the positive predictive value cannot be perfect with a low prevalence rate. Screening should be considered for those in large cities because of the low sensitivity of targeted screening and better compliance with routine screening.

HIV antibody screening should be offered to people with high-risk behaviours or those in high-risk groups because of good evidence of the effectiveness of early treatment in delaying the development of AIDS and the efficacy of aerosol pentamidine prophylaxis (A Recommendation). However, labelling is a problem, and there is no information about the long-term effects of treatment.

Cohort studies suggest that testing followed by counselling may reduce the spread of HIV infection among injection drug users and homosexual men.

There is fair evidence to recommend HIV antibody screening for neonates of HIV-positive women (B Recommendation); however, antibody screening is not specific or sensitive for infection, and other diagnostic tests, such as the viral DNA polymerase chain reaction or virus isolation, must be done. Follow-up and vaccinations will be different for seropositive children.

There is insufficient evidence to recommend the inclusion or exclusion of HIV antibody screening in low-risk populations (C Recommendation). The harm caused by false positive results must be weighed against any treatment benefits gained by the few seropositive people identified.

Unanswered Questions (Research Agenda)
Strategies to improve the identification of people who should undergo testing need to be evaluated. Although the results of initial studies of pamphlets are discouraging, the efficacy of cheaper alternatives to individual counselling should be considered.

The appropriate frequency of screening needs to be studied further.

Although public education on needle-exchange programs, counselling on condom use and counselling of seropositive people reduce high-risk behaviour they do not eliminate it. Therefore, further studies in this area are needed.

Studies of AZT and other antiviral agents, either alone or in combination, should include patient populations other than homosexual men to determine the generalizability of the results. Such studies should examine the optimal point at which treatment should be started and the long-term effects of therapy. The incremental improvement in efficacy with the use of an immunomodulator should also be studied.

Evidence
The literature was identified with a MEDLINE search to September 1991 using the following MESH headings: HIV antibody tests, AIDS, and sex-education, sex-counselling, sex-behaviour, attitude, screening, truth-disclosure, counselling, and patient education.

This review was initiated in June 1988 and recommendations were finalized by the Task Force in November 1991. A report with a full reference list was published in September 1992.<37>

Acknowledgements
The task force thanks Alfred O. Berg, MD, MPH, Professor and Associate Chair, Department of Family Medicine, University of Washington, Seattle, Washington; Abbyann Lynch, Director of Bioethics, Hospital for Sick Children, Toronto, Ontario, and Martin Schechter, MD, FRCPC, Department of Health Care and Epidemiology, University of British Columbia, Vancouver, British Columbia for reviewing the manuscript.

Full Citation
Wang E.E.L. Screening for HIV antibody. In: Canadian Task Force on the Periodic Health Examination. Canadian Guide to Clinical Preventive Health Care. Ottawa: Health Canada, 1994; 708-18.