Background: This paper evaluates the effect of the World Starts With Me (WSWM), a comprehensive sex education programme in secondary schools in Uganda. The aim of the present study was to assess the effects of WSWM on socio-cognitive determinants of safe sex behaviour (delay; condom use and non-coercive sex). Methods: A survey was conducted both before and immediately after the intervention among students in intervention (N = 853) and comparison (N = 1011) groups. A mixed model repeated measures analysis was performed to assess the effectiveness of the WSWM programme on the main socio-cognitive determinants of safe sex behaviour at post-test. A similar post-hoc comparison was made between schools based on completeness and fidelity of implementation of WSWM. Results: Significant positive effects of WSMW were found on beliefs regarding what could or could not prevent pregnancy, the perceived social norm towards delaying sexual intercourse, and the intention to delay sexual intercourse. Furthermore, significant positive effects of WSWM were found on attitudes, self-efficacy and intention towards condom use and on self-efficacy in dealing with sexual violence (pressure and force for unwanted sex). A reversed effect of intervention was found on knowledge scores relating to non-causes of HIV (petting, fondling and deep kissing). A follow-up comparison between intervention schools based on completeness of the programme implementation revealed that almost all significant positive effects disappeared for those schools that only implemented up to 7 out of 14 lessons. Another follow-up analysis on the basis of implementation fidelity showed that schools with a "partial" fidelity score yielded more significant positive effects than schools with a "full" fidelity of implementation score. Conclusions: The study showed an intervention effect on a number of socio-cognitive determinants. However, the effectiveness of WSWM could be improved by giving more systematic attention to the context in which such a programme is to be implemented. Implications for the systematic development and implementation of school-based safe sex interventions in Uganda will be discussed.