Since the development of the first detuned EIA to detect recent HIV-1 seroconversion [1], there has been a growing interest in the application of laboratory methods to measure HIV incidence in various cross-sectional populations [2]. Measurement of HIV incidence is important for monitoring effectiveness of prevention programs, targeting resources and second- generation surveillance. Until recently, laboratories had to rely on commercially available diagnostic EIA tests performed under a modified protocol to estimate HIV-1 incidence [3]. However, such detuned EIAs are mainly useful in populations with subtype B infections [4,5] and required extreme dilution of samples (e.g. 1:20,000).
The Aware™ BED™ EIA is a second-generation assay to detect recent HIV-1 seroconversion [6] and was developed to address some of the shortcomings of the detuned EIAs (e.g. assay variability, subtype-dependent performance) [4, 5]. The Aware™ BED™ EIA has been used in a number of cross-sectional populations to estimate incidence and evaluate association with various risk factors [7, 8, 14, 15, 16, 17, 18]. The HIV-1 Incidence EIA was developed and initially manufactured and distributed by the United States Centers for Disease Control (CDC). Calypte Biomedical Corporation has licensed the Aware™ BED™ EIA from the CDC and has been manufacturing and distributing the test under its previous name, “Calypte HIV-1 BED Incidence EIA”, since 2004.
The Aware™ BED™ EIA is an IgG-capture enzyme immunoassay in which the wells of a microplate are coated with goat antihuman IgG. When serum or plasma is added to the wells, anti-HIV IgG and non-anti-HIV IgG are captured on the goat-anti-human IgG-coated wells. The relative amounts of anti-HIV IgG and non-anti-HIV IgG captured represent IgG antibody populations found in the serum or plasma. Indirectly, the test measures the proportion of HIV-1-specific IgG in a given specimen with respect to total IgG. Early seroconverters have a lower proportion of HIV-specific IgG in the serum/plasma than those with long-term infection [6, 13]. Although the same specimens may have high optical density (OD) values on regular diagnostic EIAs, OD values are lower on the Aware™ BED™ EIA. Studies have indicated that HIV-specific IgG may continue to increase for more than 2 years after seroconversion when tested by this assay [6].
Original findings indicated a cutoff (ODn) of 1.0 represented a mean seroconversion duration of 160 days [6]. Further evaluations and analyses using additional panels with other subtype infections suggest that the ODn of 0.8 corresponds to mean seroconversion duration of 155 days. This cutoff of 0.8 yields few false recent infections in AIDS patients, resulting in a better predictive value for detection of recent infection.
The predictive value of any assay depends on the prevalence of that condition in a population. Therefore, the predictive value of detecting recently infected individuals in low incidence populations would be lower than in higher incidence populations.
Overall performance of the assay has been well studied examining reproducibility, inter-run and intra-run coefficient of variation (CV), and inter-operator variability [9], and data suggest that the assay has very high reproducibility with R2 > 0.9.
Classification of individuals by the Aware™ BED™ EIA as recent seroconverters or long-term infections is based on average timeframes in which individuals develop HIV antibodies, calculated from data using a large number of people [6]. However, there are differences among the individuals in the rates at which antibodies are produced. Although this assay is useful at the population level, its predictive value for the individual may be low (especially when ODn levels are close to the cutoff). Therefore, the assay should not be used for individual diagnosis.
About 2% to 3% of people with long-term HIV infections, including AIDS, may be misclassified as recently infected. Efforts should be made to exclude people that have developed AIDS or low CD4 counts to increase the predictive value of the assay.
Minor variations of the formula for calculating incidence have been used in earlier studies [1,6]. A consensus formula was agreed upon at the US CDC for calculating incidence. Annual HIV-1 incidence is calculated using the following consensus formula:

The total number of people tested, number of seronegatives, number of seropositives and number recently infected must be known for calculating incidence in a given cross-sectional population. Note that calculated incidence does not differ significantly when different formulae are used. However, a consistent approach is recommended for rational comparisons among populations and trend analysis.
The 95% confidence interval (CI) for the Incidence estimate is:

This new formula for calculating the 95% CI best fits the observed CI in statistical modeling and is dependent on the calculated incidence and the number found to be recently infected [10].
In 2006, the U.S. CDC recommended an adjusted formula that corrects for misclassification for incidence calculation and improves incidence estimation [13, 19]. The specific recommendations are available at the CDC website here.
If there is evidence of overestimation of incidence, as observed in some studies, you may seek assistance from the CDC as outlined in the recommendations.