Few studies have investigated antiretroviral (ARV) knowledge and self-efficacy in limited literacy patients. Using a randomized controlled study design, we investigated the influence of a simple pre-tested patient information leaflet (PIL) containing both text and illustrations on HIV-and ARV-related knowledge and on self-efficacy over six months in a limited literacy African population. The recruited patients were randomly allocated to either control (standard care) or intervention group (standard care plus illustrated PIL). HIV and medicines-related knowledge was evaluated with a 22-question test at baseline, one, three, and six months. Self-efficacy was assessed using a modified version of the HIV Treatment Adherence Self-Efficacy Scale. Two-thirds of the patients were female, mean age was 39.0 +/- 9.6 years and mean education was 7.3 +/- 2.8 years. Patients who received the PIL showed a significant knowledge increase over the six-month period (62.0-94.4%), with improvement at each subsequent interview whereas the control group showed no improvement. At baseline, side effect knowledge was the lowest (50-56%) but increased in the intervention group to 92%. Similarly, other medicine-related knowledge at baseline (57-67%) improved significantly (93%) and was sustained over six months. Cohen's d values post-baseline ranged between 1.36 and 2.18, indicating a large intervention effect. Self-efficacy improved significantly over six months in intervention but not control patients. At baseline, patients with <= 3 years of education had lower knowledge and self-efficacy but this was not observed post-intervention, which we attribute to the PIL mitigating the effect of limited education. Knowledge and self-efficacy were significantly correlated in the intervention group. In conclusion, a low-cost intervention of a well-designed, pre-tested, simple, illustrated PIL significantly increased both ARV knowledge and self-efficacy in HIV patients with limited education.