Discussion

Incidence and risk factors:

Widespread screening for HIV is recommended by several organizations in order to achieve goals (1) of identifying those infected, (2) getting people living with HIV (PLWH) on therapy, and (3) ultimately viral suppression. However, the incidence of HIV is not uniform throughout the population.

While the CDC gathers specific data about the incidence of HIV in the United States, determining incidence in higher risk populations such as men who have sex with men (MSM) becomes more difficult due to a lack of an accurate measure of the total MSM population in the US. Several studies have attempted to estimate the MSM population in the US and HIV incidence in this population.

The risk factor for MSM was estimated at 10 based on Purcell et al’s [2] epidemiologic meta analysis of US population studies to determine the total MSM population (utilizing a definition of MSM as a man who has sexual contact without another man in the last five years). This was then combined with CDC data to determine the HIV rate in the US MSM population, which in turn, was used to determine the likelihood ratio for HIV infection in the MSM population.

Due to the major modes of transmission being sexual contact and injection drug use (IDU), both were examined. The risk factor for various types of unprotected sex were determined by utilizing Baggaley et al’s [4] investigation of unprotective insertive and receptive anal intercourse. Patel et al’s [3] analysis was then utilized due to it’s comparison of risk for need sharing and vaginal intercourse, yielding the likelihood ratios as indicated above. Based on this data, unprotected receptive anal intercourse has been cited as carrying a risk 10-14 times greater than insertive anal intercourse at acquiring HIV, which in turn is about twice as likely to transmit HIV than vaginal intercourse.

We elected to display the most common methods of HIV transmission. While risk factors were simplified in this calculator, there are data regarding the risk as it is changed by use of barrier protection, ART use of infected individual, and viral load of infected individual which are not delineated in this calculator. Additionally, this calculator is limited as the studies used to deteremine the above likelihood ratios are meta analyses of pooled data, then processed through epidemiologic calculations lending themselves to further error and bias.

Symptoms/Syndromes:

We included symptoms both of acute HIV and chronic HIV, including opportunistic infections. A comprehensive analysis [6] of symptoms of acute HIV disease was performed and reported positive likelihood ratios for 17 different symptoms, with the most notable summarized in the table. For P. jiroveci pneumonia (PJP), a recent study [7] demonstrated that only 26% of patients with confirmed PJP pneumonia occurred in people with HIV.

Test sensitivity & specificity:

Initial testing of sensitivity and specificity of 4th generation antigen / antibody testing indicated (decreased specificity) in the setting of use in low incident populations. This may represent a difference in clinical vs analytical analysis of tests.

HIV-1 and HIV-2 Ab EIA testing is commonly used as a confirmatory test for positive 4th generation screening tests. Again, in analytical testing sensitivity reports are reported as 100%. This does not include patients who may be getting screening tests during the window period in which they are infected with HIV but have not yet developed antibodies.