What Is an HIV Test? Types, Accuracy, and Results

An HIV test detects signs of infection with human immunodeficiency virus, either by finding the virus itself or by finding the immune system’s response to it. The specific “sign” a test looks for matters a great deal: some tests hunt for antibodies your body makes against HIV, others look for a viral protein called p24 antigen, and still others search for the virus’s genetic material directly. Each approach has different strengths, different detection windows, and different roles in the testing process. What most people think of as “getting an HIV test” is usually the first step in a multi-test algorithm designed to minimize the chance of a wrong answer.

What an HIV Test Actually Detects

Your body responds to HIV in a predictable sequence after infection, and different tests are designed to intercept that sequence at different points. The virus replicates first, flooding the bloodstream with viral RNA and then with the p24 protein that forms part of its outer shell. Days to weeks later, the immune system begins producing antibodies against the virus. A test’s reliability depends on which of these targets it looks for, when those targets appear, and how concentrated they are in the specimen being tested.

This timeline is why a single test taken the day after a possible exposure can’t rule out infection. The targets simply aren’t there yet. Different test types close that gap by different amounts, which is where the concept of a “window period” comes in. We’ll get to specific window periods below, but the basic idea is straightforward: the earlier in the infection timeline a test can detect its target, the shorter the window period, and the sooner you can get a reliable result.

How HIV Tests Evolved Through Generations

HIV tests are grouped into “generations” based on what they detect and how they do it. The first HIV antibody test, introduced in 1985, was actually designed to screen blood products rather than diagnose individuals. Those first-generation tests detected only one class of antibody (IgG) and didn’t turn positive until roughly six to twelve weeks after infection.¹ That’s a long time to wait for an answer, and subsequent generations of tests have steadily closed the gap.

Second-generation tests improved on the original design by using synthetic or recombinant viral proteins instead of whole virus, which boosted specificity and reduced false positives. Third-generation tests added the ability to detect IgM antibodies, which appear earlier than IgG after infection. Some third-generation-era assays could pick up antibodies within two to four weeks of infection.²

Fourth-generation tests, now the standard in most clinical laboratories, represent the biggest jump. They detect both antibodies (HIV-1 and HIV-2) and the p24 antigen simultaneously. Because p24 antigen shows up in the blood before antibodies do, these combination tests can catch infections days earlier than antibody-only tests.³ Research comparing the timing of different test types found that the most recent antigen/antibody laboratory tests became reactive a median of about six days after direct viral RNA detection, which is significantly earlier than older antibody-only tests.⁴ The old Western blot, by contrast, didn’t become positive until about 25 days after RNA detection on average.

Rapid Tests and Point-of-Care Testing

Rapid tests give results in minutes rather than days, which makes them practical for clinics, community outreach, and home use. They come in two main forms: finger-stick blood tests and oral fluid tests. Both work, but they aren’t equally sensitive.

A systematic review comparing the two specimen types head-to-head found that oral fluid tests had a pooled sensitivity about two percentage points lower than blood-based tests. In settings where HIV is relatively common, the practical difference in positive predictive value was small. But in low-prevalence settings, the gap widened: oral fluid tests correctly identified about 89% of true positives compared to roughly 98% for blood-based rapid tests.⁵ A real-world comparison of five rapid tests in a healthcare setting confirmed the pattern: tests performed on oral fluid were consistently less sensitive than those using finger-stick blood, which were in turn less sensitive than laboratory-based serum testing.⁶

Fourth-generation rapid tests also exist. These aim to detect both p24 antigen and antibodies at the point of care, potentially catching infections during the acute phase when only antigen is present.⁷ In practice, their antigen detection component has shown mixed field performance, though their antibody detection works well. The takeaway for anyone using a rapid test: a negative result is highly reliable if you’re outside the window period, but a positive result on any rapid test should always be confirmed with follow-up laboratory testing.

Nucleic Acid Testing

Nucleic acid tests (NATs) look for HIV’s genetic material directly, which makes them the earliest detectors. They can identify infection before either p24 antigen or antibodies appear. NATs are not usually the first test someone gets because they are more expensive and take longer to process than immunoassays. Instead, they play two important roles: resolving ambiguous results from the standard testing algorithm, and screening the blood supply.

Adding pooled NAT to a testing algorithm that already uses fourth-generation antigen/antibody tests increased the detection of acute infections by about 38% in one study.⁸ That matters because people in the acute phase of infection, before antibodies develop, are among the most infectious. Catching them early has both individual and public health benefits.

NAT also serves as the confirmatory step when standard testing gives conflicting signals. In the current recommended diagnostic algorithm, a specimen that reacts on the initial fourth-generation screening test but gives a negative or unclear result on the antibody differentiation test gets sent for HIV-1 RNA testing to settle the question.⁹

How the Diagnostic Algorithm Works

No single HIV test is meant to work alone. The current recommended approach in the United States uses up to three tests in sequence. It starts with a fourth-generation combination immunoassay that screens for HIV-1 and HIV-2 antibodies plus HIV-1 p24 antigen. If that initial test is reactive, the specimen goes to a second test: an immunoassay that distinguishes between HIV-1 and HIV-2 antibodies. If the second test doesn’t agree with the first, a nucleic acid test resolves the discrepancy.¹⁰

This layered approach replaced the older method of screening with a third-generation antibody test and confirming with a Western blot. The newer algorithm catches infections earlier (because of the antigen component in the first test) and resolves ambiguous cases more definitively (because nucleic acid testing is more sensitive than Western blot during the acute phase). The tradeoff is complexity: the process can take several days if all three steps are needed, which can be an anxious wait.

Home Testing and Self-Testing

Home HIV tests let you collect a specimen and read results yourself, either with an oral fluid swab or a finger-stick blood sample. The appeal is privacy: studies consistently find that most people prefer testing in private rather than in a clinical setting.¹¹ Accuracy can be high when the test is done correctly, with one study reporting sensitivity around 97% and specificity near 100% for oral fluid self-tests when invalid results were excluded.

The catch is user error. A qualitative study observing people performing self-tests found that nearly half of participants produced invalid results because of mistakes during the process. Only about one in five completed every step without error. Finger-stick users tended to struggle most with collecting the specimen, while oral fluid users made more mistakes during the testing stage itself.¹² The lesson is that the test technology can be excellent, but real-world accuracy depends heavily on whether you follow the instructions precisely: timing the read window, using the right amount of specimen, and not eating or drinking beforehand for oral tests.

Finger-stick whole-blood home tests have generally outperformed oral fluid home tests in head-to-head comparisons. In one study, a finger-stick home test achieved 100% sensitivity across different viral load levels and CD4 counts, while oral fluid home tests ranged from about 87% to 88% sensitivity, with performance dropping further at lower viral loads.¹³ Urine-based self-tests are also emerging. A large study of a urine test kit found sensitivity above 99% and specificity of 100%, with over 93% of self-testers performing the test correctly when observed by professionals.¹⁴

Regardless of which home test you use, a positive result is preliminary. Every home test instructs you to follow up with a healthcare provider for confirmatory laboratory testing. A negative home test result is reliable only if your last possible exposure was outside the test’s window period.

False Positives and False Negatives

False positives on HIV tests are uncommon but real. Several biological conditions can trigger a reactive result on an initial screening test even when HIV is not present. These include other systemic infections that stimulate a broad immune response, certain vaccines, pregnancy, and some cancers.¹⁵ This is exactly why the diagnostic algorithm uses confirmatory tests: the initial screen casts a wide net to avoid missing any true infections, and the follow-up tests sort genuine positives from false alarms. A confirmed positive that has gone through the full algorithm is extremely reliable.

False negatives are a different concern and mostly fall into two categories. The first is testing too early, before the window period has passed. The second, rarer category involves people with severely weakened immune systems who produce very low levels of antibodies. A case report documented repeated false-negative rapid test results in a patient with advanced HIV disease whose CD4 count had dropped to just 23 cells per cubic millimeter. The severe immune suppression likely reduced antibody levels below the detection threshold of the rapid test.¹⁶ Cases like this are unusual, but they illustrate why clinicians sometimes order additional testing when symptoms strongly suggest HIV but initial results come back negative.

How PrEP and PEP Affect Test Results

If you’re taking pre-exposure prophylaxis (PrEP) to prevent HIV, or post-exposure prophylaxis (PEP) after a potential exposure, the antiretroviral drugs can complicate testing. These medications work by suppressing viral replication, which means that if you do acquire HIV while on PrEP or PEP, your viral load may be too low for some tests to detect. The drugs can suppress HIV RNA levels, potentially interfering with the nucleic acid testing step in the diagnostic algorithm.¹⁷ The antibody response may also be delayed or blunted.

This doesn’t mean PrEP users should skip testing. Regular testing is a core part of PrEP care, typically every three months. But it does mean that if you’ve recently stopped PrEP or PEP and want a definitive test, you may need to wait longer than someone who was never on these medications for the test to fully reflect your status. If you’re on PrEP and have a potential exposure event, discuss the timing and type of follow-up testing with your provider rather than relying solely on a home test.

Testing Infants Born to HIV-Positive Mothers

Standard antibody-based HIV tests don’t work the same way in newborns. During pregnancy, a mother’s antibodies cross the placenta and enter the baby’s bloodstream. If the mother is HIV-positive, her HIV antibodies will be present in her infant regardless of whether the baby is actually infected. These passively transferred antibodies can persist until the child is roughly 18 months old.¹⁸

Because of this, diagnosing or ruling out HIV in infants relies on nucleic acid testing, which looks for the virus’s genetic material rather than antibodies. A negative antibody test before 18 months is informative (it suggests the baby likely doesn’t have the infection), but a positive antibody test in that age range is ambiguous because it might just reflect mom’s antibodies. NAT-based testing is typically done at specific intervals during the first months of life to catch any infection as early as possible.

Screening the Blood Supply

The stakes for HIV testing in blood donation are exceptionally high because the consequences of a missed infection are direct: a contaminated unit could transmit the virus to a recipient. Blood banks use the most sensitive testing available, including individual-donation nucleic acid testing. In a large-scale South African screening program, individual-donation NAT caught 16 HIV-positive donations that were still in the window period and would have been missed by antibody-based screening alone.¹⁹

In the United States, nucleic acid testing of donated blood has driven the residual risk of transfusion-transmitted HIV down to roughly one in two million units.²⁰ A recent large cohort study from Eastern India confirmed that NAT remains essential for catching early infections that antibody-based and antigen-based screening methods miss.²¹ Blood supply screening is one context where the extra cost and processing time of NAT is considered non-negotiable.

Viral Diversity and Its Impact on Testing

HIV isn’t a single uniform virus. It comes in two major types (HIV-1 and HIV-2) and multiple subtypes, groups, and recombinant forms. Most tests are designed and validated primarily against HIV-1 subtype B, which is the dominant strain in North America and Western Europe. The reliability of both antibody-based and molecular tests is generally high across major HIV-1 subtypes, but the more genetically divergent variants, including HIV-1 groups O and N and HIV-2, can pose diagnostic challenges.²²

For antibody-based tests, this is less of a practical problem than it might sound, because the antibody response tends to cross-react across subtypes. One study examining HIV antibody avidity-based testing in African individuals infected with non-B subtypes found no significant association between viral subtype and test misclassification.²³ The bigger concern is with molecular tests that target specific genetic sequences: as the virus diverges from the sequences the test was designed around, sensitivity can drop. Test manufacturers continually update their assays to keep up with circulating viral diversity, but this remains an active area of surveillance.

Point-of-Care Testing in Resource-Limited Settings

Much of the world’s HIV burden is in regions where centralized laboratory infrastructure is limited. Point-of-care tests that don’t require refrigeration, electricity, or trained laboratory staff have been critical for expanding access to testing in these settings. But practical barriers remain beyond the test itself: supply chains for test kits, training for the people administering them, and the systems needed to connect someone who tests positive to treatment.²⁴

Linkage to care after a positive test result is one of the most important and underappreciated parts of the testing process. Testing that doesn’t lead to treatment has limited impact. Research has consistently shown that structured counseling and follow-up services after testing significantly improve the likelihood that a person newly diagnosed with HIV will enter and remain in care.²⁵ This is true regardless of the testing technology used. A perfect test is only useful if the person who gets a positive result has a clear path to a provider who can start treatment.