What is already known about this subject• Self‐medication is extensively practised in both developed and less‐developed countries, sometimes inappropriately.• Educational intervention in secondary schools has been proven to be useful.• Most educational interventions in adolescent populations have focused on the reduction of addictive substance abuse.What this study adds• Educational intervention can improve knowledge about self‐medication and reduce misconceptions about diarrhoea, common cold and vitamins in an adolescent population.• A specific lecture followed by small working‐group seminars produces better results than a general lecture alone in terms of ‘knowledge’ and ‘attitude’.• The positive effects of the intervention are detectable even 1 year later. Yearly reinforcing interventions while in secondary school would allow long‐lasting effects.AimImproving knowledge about rational drug use at an early age may be a good way to increase the population’s awareness of health, medicines and self‐medication. We set out to evaluate the short‐ and long‐term effects of an educational intervention to promote rational drug use and self‐medication in secondary school students.MethodsA non‐randomized, controlled clinical trial. The participants were 367 female students (10–13 years old) from two secondary public schools of the metropolitan district of Quito (Ecuador). The educational campaign had two components [a specific lecture (intervention and control schools) and subsequent small working group seminars (intervention school)] providing short and clear messages of five topics related to rational drug use. The main outcome measures were an increase in ‘knowledge’ short term (1 month) and long term (up to 1 year) after intervention and the relative risk (RR) reduction in misconceptions or wrong ideas about medicine use.ResultsThe intervention group showed a significant increase in knowledge both short and long term and in comparison with the control group, mainly regarding oral rehydration salts preparation (+59.4%; P < 0.001), lack of multivitamin energizer action (+57.4%; P < 0.001), healthy growth effects (+53.3%; P < 0.001) and the perception that medicines’ promotional activities do not teach how to take care of health (+54.0%; P < 0.001). A RR reduction in misconceptions about drugs was found short term and long term. The intervention group was less predisposed to consume antidiarrhoeals [RR = 0.75, 95% confidence interval (CI) 0.62, 0.92], cough suppressants (0.44, 95% CI 0.35, 0.55) and other medicines for the common cold (0.56, 95% CI 0.45, 0.70). Misconceptions concerning the benefits of multivitamin preparations were reduced in 73%; additionally, the intervention group showed a decrease in their consumption (43.9% basal; 25.3% short term and 25.6% long term; P < 0.001).ConclusionIt is possible to achieve a favourable modification of attitudes to appropriate use of medicines in a teenage population and this modification lasts at least 1 year. Continuous reinforcing interventions would allow better and long‐lasting effects and could help to fill the gap in health education of the general population.