We investigated factors potentially associated with the uptake of HIV voluntary counselling and testing (VCT), which is the first step in acceding to programmes for the prevention of mother-to-child transmission of HIV infection. For the period 2001–2003, we estimated the VCT uptake among the 12,252 first-time attendees of the Antenatal Clinic (ANC) at Lacor Hospital (Gulu District, North Uganda). Associations between VCT uptake and socio-demographic characteristics and reproductive history were evaluated using log binomial regression models. VCT uptake was 55.6% for the overall study period; it increased from 51.0% in 2001 to 58.6% in 2002 and 57.7% in 2003 (P <0.001). Having some education [primary versus none, adjusted prevalence proportion ratio (PPR) =1.05, 95% confidence intervals (CI): 1.00–1.10] and being unmarried (cohabitating, PPR =1.07, 95% CI: 1.03–1.10; single/widowed/divorced, PPR =1.10, 95% CI: 1.03–1.18) were significantly associated with VCT uptake. Associations of borderline significance were found for: recent change of residence, having a partner with a modern occupation, and past use of contraceptives. VCT uptake is still low in this district of North Uganda. Although some socio-demographic factors were found to have been associated with uptake, the associations were weak and not of public-health significance.