Introduction: Increasing our knowledge on geographic areas and key populations most affected by HIV is essential to improve prevention and care and to ensure a more focused HIV response. Here, we estimated the prevalence of undiagnosed HIV infections in Belgium and its distribution across geographic areas and exposure groups.
Methods: We used surveillance data on newly diagnosed HIV cases and a previously developed back-calculation model to estimate number and prevalence rates (per 10000) of undiagnosed HIV infections by exposure group at national and subnational levels. Belgium consists of three regions: Flanders, Brussels-Capital Region and Wallonia. We produced estimates for Brussels-Capital Region and Wallonia. For Flanders, we produced estimates for two sub-regional areas: the province of Antwerp and the other provinces, because Antwerp is the second largest city after Brussels. Population sizes were determined using data from the Belgian Statistical Office and surveys on sexual behaviour and drug use.
Results: In Belgium, in 2015, an estimated 2818 (95% confidence interval: 2494 to 3208) individuals were living with undiagnosed HIV, that is, 15% of individuals living with HIV. The Brussels-Capital Region and the province of Antwerp, which host the two biggest cities, accounted for ~60% of the undiagnosed infections, and had the highest undiagnosed prevalence rates per 10000: 12.0 (9.4 to 15.3) and 7.4 (5.6 to 9.8) respectively. Individuals with foreign nationality accounted for 56% of the total number of undiagnosed infections, and were the most affected populations in all areas in terms of undiagnosed prevalence rates. Specifically, men who have sex with men (MSM) with non-European nationality were the most affected population in the province of Antwerp (853.4 (408.2 to 1641.9) undiagnosed infections per 10000), the Brussels-Capital Region (543.9 (289.1 to 1019.1)), and the other provinces of Flanders (691.7 (235.5 to 1442.2)), while in Wallonia, it was heterosexual women with Sub-Saharan African nationality (132.2 (90.6 to 178.5)).
Conclusions: Geographic areas hosting the biggest cities in Belgium accounted for the vast majority of undiagnosed HIV infections and individuals with foreign nationality were the most affected, especially MSM with non-European nationality. This should be accounted for when tailoring prevention and testing programs. Furthermore, MSM with foreign nationality require more attention in Belgium, and certainly more generally in Europe.