This publication is provided for historical reference only and the information may be out of date.
For key question 1, we included randomized trials and observational studies that compared clinical outcomes in patients screened and not screened for HIV infection.
For key question 2, we included recent large U.S. observational studies reporting the prevalence of HIV in patients with different risk factors, and observational studies reporting results of risk factor assessment for targeted screening.
For key questions 3 and 4, we included studies that evaluated the diagnostic accuracy of screening tests for HIV infection and performed an appropriate reference standard on all tests. We focused on Food and Drug Administration-approved rapid HIV screening tests and included published and unpublished studies on the diagnostic accuracy of these.
For key question 5, we included recent large U.S. observational studies reporting CD4 counts or viral loads at the time of diagnosis or presentation, the proportion of patients diagnosed with HIV infection within one year of being diagnosed with AIDS, and clinical trials and observational studies reporting long-term effects of late diagnosis. We also included clinical trials and observational studies reporting uptake of voluntary HIV testing, rates of return for post-test counseling, and proportion of patients qualifying for interventions who were receiving them.
For key question 6, we included studies reporting harmful effects from performing CD4 count and HIV viral load testing in patients found to be positive, such as labeling, anxiety, and effects on close partnerships.
For key questions 7a, 7b, and 7c, we included controlled trials of interventions (highly active antiretroviral therapy [HAART], counseling, routine monitoring and follow-up, pap smears, immunizations, chemoprophylaxis for opportunistic infections) that evaluated relevant intermediate (viral load, CD4 counts, behavior changes) or clinical outcomes (clinical progression, mortality, quality of life, functional status, spread of disease) in treatment-naïve populations. We included only fully published head-to-head trials of HAART. We also included large observational studies on the effects of HAART on mortality, the effectiveness of immediate versus deferred HAART, and for interventions (such as counseling) for which there was insufficient data from clinical trials.
For key question 8, we included controlled trials and observational studies that reported adverse events from HAART in treatment-naïve populations. We focused on studies reporting risks of long-term cardiovascular harms from HAART.
For key question 9, we included randomized trials and large observational studies evaluating the relationship between changes in intermediate outcomes (viral load, CD4 count, and behavior change) and clinical outcomes (AIDS, death, spread of disease, and health-related quality of life) in patients receiving HAART and counseling.
Agency for Healthcare Research and Quality (US), Rockville (MD)
Chou R, Smits AK, Huffman LH, et al. Screening for Human Immunodeficiency Virus in Pregnant Women [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2005 Jul. (Evidence Syntheses, No. 39.) Appendix B. Inclusion / Exclusion Criteria By Key Question.