Premarital sex and non-condom use are both associated with HIV infection. This study used data from the Malawi Demographic and Health Survey (MDHS) 2004 to estimate the prevalence of premarital sexand examine the association between religion and pre-marital sex among never married Malawian men. The associations between the variables were assessed by logistic regression analysis. A total of 1039 never-married men, mean age 19.6 years (standard deviation=4.3 years) participated in the MDHS of 2004. A total of 671 (63.2%) of the men reported having ever had sex, that is, reported premarital sexual intercourse (PSI) and median age at sexual debut was 16 years. Of the 450 who answered the question whether condom was used at last sex, 213 (48.1%) reported having used condom. Compared to Catholics (as referents) the adjusted odds ratios for PSI were as follows: Presbyterians 0.83 (0.45-1. 53),
Seventh-Day Adventists (SDAs) or /Seventh-Day Baptists (SDBs), 0.82 (0.35-1.90), Other Christians 1.00 (0.63-1.60), Muslims 1.96 (0.96-4.00), those with no religion 1.95 (0.37-3.549). The adjusted odds ratios for condom use at last sex were: Presbyterians 1.05 (0.53-2.09); SDAs/SDBs 1.10 (0.41-2.95); Other Christians 0.90 (0.47-1.71); Muslims 1.02 (0.41-2.54); No religion 0.56 (0.08-3.77). There was no overall difference on the likelihood of reporting having ever had premarital sex among never-married men among who reported to belong to a religious group or those not belonging to a religious group. Although individuals who reported not belonging to any religion were almost half as likely to have used condoms at last sex, the effect estimate was not significant. Reporting no religion was not associated with premarital sex nor condom use among men in Malawi.