In the Democratic Republic of Congo no previous studies have assessed the factors associated with different patterns of condom use and with multiple sexual partners, and the association between condom use simultaneously taking into account multiple sexual partnerships, and HIV infection. We carried out a prospective case-control study. From December 2010 until June 2012, 1630 participants aged 15-49 getting HIV Voluntary Counseling and Testing in a hospital in Kinshasa were selected. Cases were new HIV diagnosis and controls were HIV-negative participants detected along the study period. We recruited 274 cases and 1340 controls that were interviewed about HIV-related knowledge, attitudes and behaviours. Among cases there was a high prevalence of multiple lifetime and concurrent sexual partnerships (89.8% and 20.4%, respectively) and most cases never used condoms with only 1.5% using them consistently. Condom use and multiple partnerships were associated with male, single and high-educated participants. An association was found between multiple lifetime partners and 'any condom use' (OR = 2.99; 95%CI: 2.14-4.19) but not with consistent use. Both having two or more multiple concurrent sexual partners or not using condoms were variables similarly and highly associated to HIV risk. The association found between having two or more concurrent sexual partners and HIV was slightly higher (OR = 3.58, 95%CI:2.31-5.56) than the association found between never condom use and HIV (OR = 3.38, 95%CI:1.15-9.93). We found a high prevalence of multiple lifetime sexual partners and an extremely high prevalence of inconsistent condom use, both strongly associated with HIV seropositivity. Local programmes would benefit from comprehensive interventions targeting all behavioural and sociocultural determinants.