Modelling the impact of serosorting and seroadaptation on the spread of HIV in men who have sex with men

Serosorting - the practice of choosing sexual partners based on their perceived serostatus - is widely credited as a behavioural intervention that limits the transmission of HIV among men who have sex with men (MSM). However, if this assumption is false, the trend towards serosorting could potentially promote the spread of HIV infection. Here we present a deterministic compartmental model of ordinary differential equations and a subsequent network model of HIV transmission among an MSM population to study the impact of serosorting on HIV incidence and prevalence. Analysis of the compartmental model suggest that serosorting is an effective preventive measure at the population level only once a critical mean time to diagnosis has been achieved. The detrimental impacts of serosorting associated with longer times to diagnosis in the compartmental model are nearly eliminated in the subsequent network model, demonstrating the importance of considering network structure in models of this kind.