The methods and limitations for estimating HIV incidence vary based on the data and surveillance systems available in countries.

Countries with high HIV prevalence in the general population have relatively strong surveillance systems with household surveys contributing to the information required to estimate incidence. In epidemics concentrated in key populations, the surveillance systems for key hard-to-reach populations are often not comparable over time due to changing survey and sampling methods. The estimated size of key populations, a critical input to the Spectrum model for concentrated epidemics, can also lead to important under or over estimation of HIV incidence in concentrated epidemics.
In many countries trends in recent new infections rely on prevalence data from routine antenatal clinic testing. If those data are biased because women with known positive HIV status are not captured when calculating prevalence, or women found to be negative at initial antenatal care visit are retested later in the pregnancy, the derived incidence trends might be biased. While some limitations of the models are reflected in the uncertainty bounds the measurement biases and the uncertainty caused by these biases are not easily quantified and are thus not included.
Although HIV prevalence and incidence among children appears to be reasonably robust in generalized epidemics, estimating the pediatric HIV epidemic in concentrated epidemics remains a challenge because no robust measures of fertility exist among key populations living with HIV.
Currently UNAIDS only supports the HIV estimates development in countries with populations greater than 250,000. This is primarily due to support capacity at UNAIDS.

Hesin internet-kagin er ikki longur stuðlaður

Rætta uppseting: 3.3.1