Anecdotal reports have suggested that reduced efficacy of pediculicides against Pediculus humanus capitis could be related to resistance to treatments. Ovicidal and pediculicidal activities of 0.5% malathion and 0.3% d-phenothrin lotions were tested in an experimental model of P humanus capitis grown on rabbits to ensure that the two treatments were pharmacologically equipotent. We then did a randomised controlled trial in which the lotions were administered to 193 P humanus capitis-infested schoolchildren (malathion, 95; d-phenothrin, 98). Success rate was defined as the absence of both live lice and viable nits. Before treatment, live lice were collected and subjected to a pediculicidal test. Pharmacological tests showed 100% killing of the rabbit-grown nits and lice after exposure to both pediculicides. On day 1 of the controlled trial, the success rate was 92% in the malathion group (95% CI, 0.86-0.97) and 40% in the d-phenothrin group (0.30-0.49) (p < 0.0001); on day 7, it was 95% in the malathion group (0.90-0.99) and 39% in the d-phenothrin group (0.29-0.48, p < 0.0001). Malathion was also significantly more active in the pediculicidal tests compared to d-phenothrin and control. These results suggest an acquired resistance to d-phenothrin in the schoolchildren tested, since all other conditions of the administration of insecticides were standardised.