Early HIV antibody tests used antibodies that bound to the non-variable parts of human antibodies. These antibodies were then linked to an enzyme - another protein - that drives an easily-detected chemical reaction. The combined or 'conjugated' protein is then manufactured in large quantities.

An enzyme called horseradish peroxidase is commonly used because it is cheap and easy to link to other proteins. It drives a simple chemical reaction that is easy to see and to measure because it changes the colour of a test solution.

One alternative to enzyme-linkage is to label antibodies with a substance called fluorescein which glows bright green under UV light. This is the basis for IFA tests (which are explained later in this section).

A limitation of the earliest HIV antibody tests was that they used antibodies that only bound to IgG antibodies. Of the five classes, IgG is by far the commonest in blood. However, the very first antibodies in the blood of someone who is HIV-positive are often of the IgM class.

The more modern tests, as we shall see, can recognise IgM as well as IgG, enabling detection of antibodies approximately one week earlier in the course of HIV infection.

Mothers and babies

A mother's IgG antibodies cross the placenta during pregnancy into her baby, but not her IgA antibodies. Tests for IgA have therefore been used to identify HIV-positive babies. However, these tests don't work very well in the youngest babies. If viral load tests are available, as they are in the UK, they offer a better alternative (though it is best to make sure that the particular test works for the mother's strain of the virus, before relying on it to tell whether her baby is HIV- positive).