Share this article on:

Selecting an HIV Test: A Narrative Review for Clinicians and Researchers

Hurt, Christopher B. MD*†; Nelson, Julie A.E. PhD*‡; Hightow-Weidman, Lisa B. MD, MPH*†; Miller, William C. MD, PhD, MPH§

doi: 10.1097/OLQ.0000000000000719

Abstract: Given the many options available, selecting an HIV test for a particular clinical or research setting can be daunting. Making an informed decision requires an assessment of the likelihood of acute infection in the test population and an understanding of key aspects of the tests themselves. The ability of individual tests to reliably detect HIV infection depends on the target(s) being detected, when they can be expected to be present after infection, and the concentration of stable target in test specimens, all of which are explained by the virologic and serologic events after infection. The purpose of this article is to review the timeline of HIV infection, nomenclature, and characteristics of different tests; compare point-of-care and laboratory-based tests; discuss the impact of different specimens on test performance; and provide practical advice to help clinicians and researchers new to the field select a test that best suits their needs.

This narrative review summarizes current testing technologies and discusses key considerations in selecting appropriate tests for different clinical and research applications.

From the *Institute for Global Health and Infectious Diseases, †Behavior and Technology Lab, and ‡Department of Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, NC; and §Division of Epidemiology, College of Public Health, Ohio State University, Columbus, OH

On the basis of a presentation given by C.B.H. at a joint consultation on “Enhancing the Measurement of HIV Exposure Due to Sexual Behavior” for the National Institute of Mental Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development, and National Institute of Allergy and Infectious Disease on March 22, 2016.

Conflict of Interest and Sources of Funding: C.B.H. is supported by the National Institute of Mental Health (K23MH099941). J.A.E.N. is supported by the University of North Carolina at Chapel Hill Center for AIDS Research, a National Institutes of Health–funded program (P30AI50410). L.H.W. is supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R43HD088319). W.C.M. was supported by the National Institute of Allergy and Infectious Diseases (R01AI114320) and Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01HD080485). The views expressed do not necessarily reflect those of the National Institutes of Health.

Correspondence: Christopher B. Hurt, MD, Institute for Global Health and Infectious Diseases, University of North Carolina at Chapel Hill, 130 Mason Farm Rd, CB #7030, Chapel Hill, NC 27599-7030. E-mail:

Received for publication April 3, 2017, and accepted July 23, 2017.

Accurate knowledge of one's HIV status remains the most important element in strategies to prevent HIV infection. Diagnosed persons can be linked to life-sustaining antiretroviral therapy that also reduces their likelihood of transmitting HIV to others, whereas at-risk persons can leverage a variety of behavioral and biomedical tools, such as preexposure prophylaxis (PrEP), to remain uninfected.

In the United States, recent Centers for Disease Control and Prevention (CDC) surveillance data suggest a stabilization or a decrease in HIV incidence1 in the face of apparent increases in the numbers of persons tested among key risk groups.2 The reasons for these improving trends are not yet clear, but to be sustained, we must continue to refine systems for HIV testing and linking persons to care and prevention resources, as appropriate.

Decisions about which HIV test to use are often made behind the scenes by laboratory professionals, leaving many clinicians and researchers unfamiliar with important tests they routinely use. The purpose of this article is to provide an overview of current HIV diagnostic technologies and recent changes in testing nomenclature, to describe how test performance varies with stage of infection and type of specimen tested, and to help individuals better understand the advantages and disadvantages of different tests, depending on their particular practice or research setting.

Back to Top | Article Outline


Time to reactivity for any given assay depends on the following: (1) the target being detected, (2) when the target is present after infection, (3) the concentration of target in the specimen, (4) the volume of specimen tested, and (5) the test's lower limit of target detection. Because the virologic and serologic events after HIV infection determine when various targets become detectable,3,4 understanding this timeline is essential for understanding the limitations of different tests across clinical and research settings.

After an exposure that leads to infection, there is a variable amount of time called the eclipse period in which no existing diagnostic test is capable of detecting HIV (Fig. 1). HIV RNA is the first reliable marker of infection; 50% of infected individuals have detectable plasma RNA within 12 days,4 and levels peak between 20 and 30 days.5 By day 15, the HIV-1 capsid protein p24 reaches detectable levels in the plasma.5 Antigenemia with p24 continues to rise through days 25 to 30, at which point early anti-HIV antibodies are able to complex with circulating p24; by day 50, antigen is often cleared from the bloodstream entirely.7 This short-lived detectability of p24 is therefore helpful in determining recency of infection, but it also makes its utility in diagnosis time limited.

As with other infections, the serum antibody response starts with immunoglobulin class M (IgM) molecules, beginning to rise around day 20, peaking within another 10 to 15 days, and then shifting to a more mature immunoglobulin class G (IgG) response beginning between days 30 and 35.6

The time from infection to the first reactive result on a given test is referred to as the window period, the length of which depends on the target being detected by a particular assay. For example, tests capable of identifying anti-HIV IgM have a shorter window period compared with those detecting only IgG. Variability in the time needed for targets to rise to detectable levels in different persons means that the window period actually reflects a distribution of time rather than a fixed length that is identical for everyone (Table 1).

Back to Top | Article Outline


Since the US Food and Drug Administration (FDA) approved the first HIV diagnostic test in 1985, 4 additional “generations” of antibody tests for HIV have been developed; each improves incrementally on its predecessor(s) in terms of performance and shortening of the window period3,8 (Fig. 2). The official nomenclature for HIV testing is currently changing, with categorizations based on analytic target(s) instead of generation numbers.9 For example, first- and second-generation assays are now referred to as “IgG sensitive,” because they are capable of detecting only IgG. Each category of tests may be further subcategorized as point of care (POC) or laboratory-based (Table 2).

In IgG-sensitive tests, anti-HIV antibodies from patient specimens bind to recombinant or synthetic HIV antigens immobilized on the solid phase of the assay (typically gp41 for HIV-1 and gp36 for HIV-2), after which a detection reagent is applied. For POC tests, the detection reagent is often a conjugate of protein A (an immunoglobulin-avid, recombinant bacterial peptide) with colloidal gold or selenium.3 The increased purity of synthetic antigens contributed to a shortening of the median window period from 45 days for viral lysate assays (such as the Western blot)10 to 31 days with current IgG-sensitive POC tests.4 The antibody-based supplemental test (which has replaced the Western blot in the US testing algorithm8) is a specialized, rapid, IgG-sensitive laboratory assay featuring multiple recombinant antigens from both HIV-1 and HIV-2. This allows for confirmation of the presence of anti-HIV IgG and differentiation of HIV-1 from HIV-2, all within 30 minutes.

IgM/IgG-sensitive (formerly third-generation) tests shorten the window period to the earliest threshold of IgM detection, a median of 23 days after infection.4 As with IgG-sensitive assays, solid-phase antigens bind antibody from the patient specimen, but instead of labeled protein A applied for detection purposes, a second, labeled or enzyme-linked synthetic HIV antigen is used, in a method referred to as an “antigen sandwich.”

Antigen/antibody (Ag/Ab) combination (formerly fourth-generation) tests pair an IgM/IgG-sensitive antibody test with simultaneous, separate p24 antigen detection. Antigen from the patient specimen is first captured by immobilized, anti-p24 antibody on the solid phase of the assay, and then a separate, labeled anti-p24 antibody is applied, forming an “antibody sandwich.” Some of these p24/IgM/IgG-sensitive tests report a reactive result if any element is detected, whereas others yield separate results for p24, anti–HIV-1 antibodies, and anti–HIV-2 antibodies. Detecting p24 shortens the median window period down to just 18 days after infection.4

In contrast to complex, automated laboratory-based platforms, POC tests rely on 1 of 2 methods: lateral flow, in which the specimen is drawn through an antigen-impregnated strip by capillary action, or flow-through, in which the patient's specimen and reagents are sequentially applied to a membrane embedded with HIV antigens (Fig. 3). Detection is accomplished most often with protein A–colloidal gold conjugates, either rehydrated from within the lateral flow strip or added separately to flow-through assays.

Back to Top | Article Outline


Laboratory-based serum or plasma assays generally offer higher sensitivities but require venipuncture, larger specimen volumes, processing, and skilled technicians. Point-of-care tests are attractive alternatives for many applications, but performance differs substantially depending on the specimen type; tests using oral transudate are significantly less sensitive than those using whole blood, and tests using whole blood are less sensitive than those using serum or plasma.11–14 This hierarchy is at least partially explained by the relative concentration of detectable target(s) per volume of specimen tested. Forty-five percent of whole blood is made up of cells, so additional processing to yield plasma or serum further concentrates antigens and antibodies per unit volume.

Compared with plasma, oral transudate immunoglobulin concentrations are 300-fold lower.15 Because oral antibodies rise to detectable levels later than those in blood, the window period is prolonged when oral fluid is used for testing. This was demonstrated most clearly in a study of HIV seroconverters in Nigeria, in which antibodies in oral transudate became detectable a median of 29 days after those in plasma.16

Back to Top | Article Outline


The introduction of the first FDA-approved rapid test in 2002 and changes in regulations permitting use of certain tests outside laboratory settings revolutionized the process of getting tested for HIV.3 Point-of-care tests are portable and easy to use, and the ability to perform HIV testing on oral transudate or fingerstick whole blood eliminates the need for venipuncture and the handling, processing, or storage of blood. This can make the entire encounter self-contained, allowing for anonymous testing where permitted by law. Point-of-care tests also greatly increase the likelihood that the patient receives his or her result. These tests have one key disadvantage, however: although the 8 current FDA-approved POC tests have excellent and generally comparable performance characteristics (Table 2), laboratory-based tests have greater sensitivity, especially early after infection.

Indeed, acute HIV-1 infection (AHI) remains a critical blind spot for current POC tests. Because AHI is characterized by the presence of p24 or HIV-1 RNA in the absence of detectable anti–HIV-1 antibodies, antibody-only tests cannot diagnose AHI. Multiple laboratory-based p24/IgM/IgG platforms are available (Table 2), but development of a POC version with comparable performance has proved more challenging. The first such test was approved by the FDA in 2013 (Determine HIV-1/2 Ag/Ab Combo; Alere, Stockport, UK). Field evaluations have demonstrated excellent sensitivity for antibody detection but variable, suboptimal sensitivity for p24.17 Studies of a newer version of this same test suggest interim improvements, with antigen sensitivity as high as 88%.14,18 Using plasma instead of whole blood enhances test performance, suggesting a role for this test in situations when laboratory support is available but an automated platform is not.14

Back to Top | Article Outline


Data are accumulating that users of tenofovir disoproxil fumarate (TDF)–based PrEP who become infected in the context of poor adherence may have delayed seroconversion on blood-based and oral POC tests. In the Partners PrEP Study, participants underwent POC testing on whole blood with both IgM/IgG and p24/IgM/IgG-sensitive tests and had samples sent for multiple laboratory-based assays every 3 months.19 Delayed seroconversion was defined as more than 100 days between the first laboratory-based evidence of infection and the first POC test reactivity for the participant. Compared with those on placebo, participants with any pharmacologic evidence of adherence during the seroconversion period had 7.2 times the odds of delayed seroconversion; plasma RNA levels were lower, as well, but still detectable.19 Among participants randomized to receive single-agent TDF in the Bangkok Tenofovir Study who subsequently became HIV infected, oral antibody reactivity trailed the first evidence of HIV (antibody or RNA) in blood by a median of 126.5 days.20 Taken together, these findings underscore the importance of using laboratory-based serum or plasma tests whenever possible in the management of PrEP users and having a low threshold for augmenting p24/IgM/IgG-sensitive assays with nucleic acid testing (NAT) among users with known or suspected poor adherence.

Back to Top | Article Outline


All HIV diagnostic tests are guided by a common principle: screen with a highly sensitive initial test and confirm reactive results with a different test that is both sensitive and highly specific. This can be accomplished using 2 POC tests, 2 laboratory-based assays, or combinations thereof; all of these strategies have been studied extensively.

In areas of the world without routine access to laboratory services,21 POC test–based strategies are the standard for diagnosis, and 2 approaches have been examined.22 Serial testing more closely approximates a “traditional” screening and confirmation approach by combining 2 different POC tests in sequence; if an initial result is reactive, a second test is run. Because discordant results may indicate recent infection, guidelines recommend either retesting in several weeks or referral for laboratory-based testing.21 In parallel testing, 2 different POC tests are run simultaneously, with a separate, third test used as a tiebreaker for discrepancies. In 2004, the World Health Organization recommended serial testing for resource-limited settings, provided that the sensitivity and specificity of tests selected are at least 99%.21

Although 2-step testing with POC assays is an acceptable, evidence-based option for HIV diagnosis in the United States,23 the majority of testing ordered in clinical encounters is processed according to the 2014 algorithm for laboratory-based HIV diagnosis (Fig. 4).8 Any reactive initial Ag/Ab combination test is confirmed using an IgG-sensitive HIV-1/2 antibody differentiation supplemental assay, a highly specific, faster, and less labor-intensive replacement for the Western blot.24 Indeterminate or negative supplemental assay results must be confirmed using NAT; the presence of HIV-1 RNA distinguishes a false-positive screening test from a likely case of AHI. Although in practice, any qualitative or quantitative HIV-1 RNA assay could be used for diagnosis (contingent on local policies and laboratory validations), only one is currently FDA approved for that purpose.3

Back to Top | Article Outline


Selecting an HIV test for a given situation requires balancing client characteristics and programmatic concerns against test performance (i.e., its sensitivity and specificity with different specimen types, its reliability in detecting AHI, and its ability to identify HIV-2 infections).25 At the client level, variation in individual preferences and testing patterns comes into play. Some individuals may be less willing to provide blood samples or to return at a later date for a test result, both of which have implications for selecting a POC test over a laboratory-based platform. Program-level considerations include building or maintaining laboratory capacity, training and supervising personnel, assuring quality, following up with clients, complying with local public health reporting requirements, and budgeting. With this framework and the data presented in mind, let us consider 4 specific situations: testing after an exposure event, screening populations at high risk for infection, outreach testing in developed countries, and selection of HIV tests for global health research.

Back to Top | Article Outline

Testing After an Exposure Event

HIV testing is often prompted by a defined exposure, such as a needlestick injury, condom failure, or condomless sex. Because no diagnostic test is capable of detecting infection in the eclipse period, baseline testing is obtained to rule out undiagnosed, preexisting infection in the patient. Blood-based POC testing should be used whenever possible, to facilitate prompt initiation of postexposure prophylaxis (if indicated).26,27 Strong preference should be given to an Ag/Ab combination test over antibody-only options, and oral fluid should not be used. Follow-up testing should occur 4 to 6 weeks and 3 months after baseline, with additional testing at 6 months if the exposure event resulted in hepatitis C virus infection.27 These recommended time points reflect expert opinion rather than a strict adherence to window period durations. If the patient tests negative at the end of the window period (Table 1), one can be very confident that he or she did not acquire HIV from the exposure.

Back to Top | Article Outline

Screening High-Incidence Populations

Point-of-care and laboratory-based tests perform similarly well for the detection of established, chronic HIV infections, provided that blood-based specimens are used. (Although rare, false-negative oral POC tests have been reported in patients with advanced HIV disease.28,29) In populations with a greater likelihood of recent infection (including persons on TDF-based PrEP), maximizing sensitivity to detect AHI is critical. To accomplish this, laboratory-based p24/IgM/IgG-sensitive assays on serum or plasma should be used whenever possible. In settings with support for specimen processing but without an automated test platform, using serum or plasma with a p24/IgM/IgG-sensitive POC test will improve its sensitivity (compared with whole blood)11,13,14 and offers the potential for earlier diagnosis compared with IgM/IgG-sensitive tests. Although most people with symptomatic HIV-1 acute retroviral syndrome (e.g., fever, pharyngitis, and lymphadenopathy) will have detectable p24 antigenemia, adding a NAT to the diagnostic workup (where available) may help capture the small proportion of those with recent infections who are RNA positive but antigen negative. (This situation may occur either very soon after the eclipse period has ended or in a rare, second window period after p24 has cleared but before IgM has become detectable.30)

Back to Top | Article Outline

Outreach Testing in Developed Countries

Mobile testing units, syringe exchange programs, and community testing events place a premium on test portability, ease of administration, flexibility in specimen type, and rapid turnaround of results, all of which make POC tests an obvious, frequent choice. However, longer window periods (Table 1) and lower sensitivities (Table 2) with POC assays compared with their laboratory-based counterparts are important trade-offs to consider. Because some outreach testing populations may be more likely to have recent HIV infection, IgM/IgG-sensitive POC tests (including p24/IgM/IgG-sensitive assays) should be prioritized over those capable of detecting IgG only. Whenever logistically feasible, serum or plasma should be tested instead of whole blood, and consideration should be given to sending a specimen for laboratory platform testing in addition to the POC test. Oral transudate should be used as a specimen only when obtaining blood (through venipuncture or fingerstick) is impractical or undesirable.13 Patients and testing clients considering oral HIV self-testing should be counseled about its longer window period compared with either blood-based POC tests or laboratory-based assays.

Back to Top | Article Outline

Selection of Tests for Global Health Research

Investigators new to HIV testing and prevention may wish to incorporate screening as part of study designs in resource-limited areas internationally. The first step is to familiarize oneself with local testing standards and approved options. As much as possible, serum or plasma should be used as the test specimen instead of whole blood to help maximize test sensitivity, and preference should be given to using nationally approved IgM/IgG-sensitive POC tests (including Ag/Ab combination assays) over IgG-sensitive tests. For most jurisdictions, reactive results from 2 sequential (and different) POC tests can be used for diagnosis, in accordance with World Health Organization guidelines.21 One can be confident in making a diagnosis of HIV from concordant results in a serial POC testing algorithm if the prevalence of infection in the test population is high (≥5%), provided that highly sensitive and specific (both ≥99%) POC tests were used. In lower-prevalence populations, the false-positive rate will always be higher with a highly sensitive test; laboratory-based confirmatory testing may help prevent unnecessary HIV care, although it is not always possible to obtain. Finally, in areas with a significant prevalence of HIV-2 or subtype O HIV-1 (primarily in western Africa), the selected tests must be capable of reliably detecting those viruses with adequate sensitivity and specificity.

Back to Top | Article Outline


The variety and complexity of available HIV test platforms can be daunting for clinicians or researchers trying to decide on the right test for their particular programmatic needs. By taking time to understand the timeline of virologic and serologic events after infection, the context of test use, how the various tests work, and the advantages and disadvantages of POC tests versus laboratory-based assays, it is possible to make better, more informed choices that improve one's confidence in the outcome of HIV testing.

Back to Top | Article Outline


1. Hall HI, Song R, Tang T, et al. HIV trends in the United States: Diagnoses and estimated incidence. JMIR Public Health Surveill 2017; 3:e8.
2. Seth P, Walker T, Figueroa A. CDC-funded HIV testing, HIV positivity, and linkage to HIV medical care in non–health care settings among young men who have sex with men (YMSM) in the United States. AIDS Care 2016:1–5.
3. Branson BM. HIV testing updates and challenges: when regulatory caution and public health imperatives collide. Curr HIV/AIDS Rep 2015; 12:117–126.
4. Delaney KP, Hanson DL, Masciotra S, et al. Time until emergence of HIV test reactivity following infection with HIV-1: Implications for interpreting test results and retesting after exposure. Clin Infect Dis 2017; 64:53–59.
5. Busch MP, Satten GA. Time course of viremia and antibody seroconversion following human immunodeficiency virus exposure. Am J Med 1997; 102:117–124; discussion 125–116.
6. Cooper DA, Imrie AA, Penny R. Antibody response to human immunodeficiency virus after primary infection. J Infect Dis 1987; 155:1113–1118.
7. Henrard DR, Daar E, Farzadegan H, et al. Virologic and immunologic characterization of symptomatic and asymptomatic primary HIV-1 infection. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 9:305–310.
8. Branson BM, Owen SM, Wesolowski LG, et al. Laboratory Testing for the Diagnosis of HIV Infection: Updated Recommendations. 2014. Available at:
9. Wesolowski LG, Parker MM, Delaney KP, et al. Highlights from the 2016 HIV diagnostics conference: The new landscape of HIV testing in laboratories, public health programs and clinical practice. J Clin Virol 2017; 91:63–68.
10. Petersen LR, Satten GA, Dodd R, et al. Duration of time from onset of human immunodeficiency virus type 1 infectiousness to development of detectable antibody. The HIV Seroconversion Study Group. Transfusion 1994; 34:283–289.
11. Pavie J, Rachline A, Loze B, et al. Sensitivity of five rapid HIV tests on oral fluid or finger-stick whole blood: A real-time comparison in a healthcare setting. PLoS One 2010; 5:e11581.
12. Pilcher CD, Louie B, Facente S, et al. Performance of rapid point-of-care and laboratory tests for acute and established HIV infection in San Francisco. PLoS One 2013; 8:e80629.
13. Stekler JD, O'Neal JD, Lane A, et al. Relative accuracy of serum, whole blood, and oral fluid HIV tests among Seattle men who have sex with men. J Clin Virol 2013; 58 Suppl 1:e119–e122.
14. Masciotra S, Luo W, Westheimer E, et al. Performance evaluation of the FDA-approved Determine HIV-1/2 Ag/Ab Combo assay using plasma and whole blood specimens. J Clin Virol 2017.
15. Mortimer PP, Parry JV. Detection of antibody to HIV in saliva: A brief review. Clin Diagn Virol 1994; 2:231–243.
16. Luo W, Masciotra S, Delaney KP, et al. Comparison of HIV oral fluid and plasma antibody results during early infection in a longitudinal Nigerian cohort. J Clin Virol 2013; 58 Suppl 1:e113–e118.
17. Conway DP, Holt M, McNulty A, et al. Multi-centre evaluation of the Determine HIV Combo assay when used for point of care testing in a high risk clinic-based population. PLoS One 2014; 9:e94062.
18. Fitzgerald N, Cross M, O'Shea S, et al. Diagnosing acute HIV infection at point of care: A retrospective analysis of the sensitivity and specificity of a fourth-generation point-of-care test for detection of HIV core protein p24. Sex Transm Infect 2017; 93:100–101.
19. Donnell D, Ramos E, Celum C, et al. The effect of oral pre-exposure prophylaxis on the progression of HIV-1 seroconversion. AIDS 2017.
20. Curlin ME, Gvetadze R, Leelawiwat W, et al. Analysis of false-negative HIV rapid tests performed on oral fluid in three international clinical research studies. Clin Infect Dis 2017; 64:1663–1669.
21. World Health Organization. Rapid HIV tests: Guidelines for use in HIV testing and counseling services in resource-constrained settings. 2004. Available at:
22. Wright RJ, Stringer JS. Rapid testing strategies for HIV-1 serodiagnosis in high-prevalence African settings. Am J Prev Med 2004; 27:42–48.
23. Centers for Disease Control and Prevention. Implementing HIV Testing in Nonclinical Settings: A Guide for HIV Testing Providers. Atlanta: US Department of Health and Human Services, 2016.
24. Centers for Disease Control and Prevention. Technical Update on HIV-1/2 Differentiation Assays. Atlanta: US Department of Health and Human Services, 2016. Available at:
25. Centers for Disease Control and Prevention. Planning and Implementing HIV Testing and Linkage Programs in Non-Clinical Settings: A Guide for Program Managers. Atlanta: US Department of Health and Human Services, 2014.
26. Kuhar DT, Henderson DK, Struble KA, et al. Updated US Public Health Service Guidelines for the management of occupational exposures to human immunodeficiency virus and recommendations for postexposure prophylaxis. Infect Control Hosp Epidemiol 2013; 34:875–892.
27. Dominguez KL, Smith DK, Thomas V, et al. Updated guidelines for antiretroviral postexposure prophylaxis after sexual, injection drug use, or other nonoccupational exposure to HIV—United States, 2016. 2016. Available at: Accessed July 12, 2016.
28. Brown P, Merline JR, Levine D, et al. Repeatedly false-negative rapid HIV test results in a patient with undiagnosed advanced AIDS. Ann Intern Med 2008; 149:71–72.
29. Stekler JD, Swenson PD, Coombs RW, et al. HIV testing in a high-incidence population: Is antibody testing alone good enough? Clin Infect Dis 2009; 49:444–453.
30. George CR, Robertson PW, Lusk MJ, et al. Prolonged second diagnostic window for human immunodeficiency virus type 1 in a fourth-generation immunoassay: Are alternative testing strategies required? J Clin Microbiol 2014; 52:4105–4108.
© Copyright 2017 American Sexually Transmitted Diseases Association