Scenario-independent parameters of the simulations.

aThe studies in [108]–[110], [112], [114] surveyed several African cities: we used the estimates provided for Yaoundé, which was the location nearest to Kinshasa.
bThese parameters determined means for random distributions, rather than fixed values.
cBoth combinations of 3 partners per year with 52-week short links and 4.5 partners per year with 26-week short links yielded a concurrency index around 1.5, consistent with [108].
dEmploying the same value as for stable links.
eThe line between CSWs and femmes libres is blurred. We therefore ran also simulations without professional CSWs, and with CSWs with intermediate sexual promiscuity.
fThere were 14 CSWs per 1,000 men in Yaoundé according to [114]. With 600 acts per CSW per year, this would yield about 8 CSW visits per man per year.
gThis formula implies the same number of concurrent links (including the spousal link) but fewer annual partners for married compared with single men.
h(GU) indicates a partner with an active genital ulcer; (C) indicates a circumcised man; (NC) indicates a non-circumcised man.