The occurrence of high rates of alcohol consumption in a context of high HIV prevalence in South Africa poses a significant health challenge for this country. This paper aims to answer three questions that could further our knowledge regarding the links between alcohol use and HIV infection: (a) "Are problem drinkers more likely to have multiple concurrent partners than those who are not?"; (b) "Are condoms applied less effectively and less consistently by problem drinkers compared to those who are not?"; (c) "Are the female sexual partners of problem drinkers different from those who are not?" Two cross-sectional HIV bio-behavioural surveillance surveys using Respondent-Driven Sampling were conducted in two peri-urban settings on the outskirts of Cape Town, South Africa. Eight hundred and forty-eight men aged 25-55 years who have multiple, concurrent female sexual partners were recruited. Problem drinkers had a score of ≥3 on the CAGE questionnaire. Questions enquired about partner numbers, condom use and partner traits. Multivariate logistic regression models were developed to determine significant associations between outcome variables and problem drinking. Fifty-eight percent of men were problem drinkers. Compared to non-problem drinkers, problem drinkers were significantly more likely to report having any symptom of a STI; not using condoms due to drinking; inconsistent condom use with all partner types; that their most recent once-off partner was unemployed; having met their most recent partner at an alcohol-serving venue; and having had a once-off sexual relationship. Alcohol may fuel once-off sexual encounters, often characterised by transactional sex and women's limited authority to negotiate sex and condom use; factors that can facilitate transmission of HIV. HIV prevention interventions specifically targeting drinkers, the contexts in which problem drinking occurs and multiple sexual partnering are urgently needed.