OBJECTIVE: Most studies of different school and community based programs aimed at preventing unwanted pregnancies have shown little effect in reducing teen age sexual risk behaviour, while a few have shown some effect. We wanted to investigate the effect of different types of sexual education programs in the junior high schools of Oslo on sexual debut, use of contraception at last intercourse, and pregnancy.METHODS: A longitudinal design was used. The base-line study in 2001 had question formulas given to all 10th graders in Oslo. The schools answered questions about the use of external sexual health educators. In the follow-up study three years later the same pupils were asked about debut, contraceptives and pregnancies. Adjusted odds ratio (ORadj) and 95 % confidence intervals (CI) were calculated using binary logistic regression analysis in the statistical program SPSS, version 14.0.RESULTS: The only significant results was a higher probability of answering yes on the question of sexual debut among girls having received education by Clinic for Sexual Health (Klinikk for seksuell helse) (ORadj = 2.8; CI: 1.55 - 5.03) or Youth Health Station (Helsestasjon for ungdom) (ORadj = 3.1;, CI: 1.33-5.11).CONCLUSION: Two of the most common institutions for sexual education in Oslo seem to increase the probability among girls to have had their sexual debut before the age of 18-19. This is in contrast to findings from most studies and should be further investigated, preferably with experimental epidemiological designs. External education seems to have no effect on use of contraceptives at last intercourse, and the low number of pregnancies makes it impossible to conclude on this outcome.