[The evaluation of the critical levels of malaria transmission in a stable endemic area (author's transl)].

During a longitudinal survey done in the degraded forest area south of Brazzaville (People's Republic of the Congo), it appeared that plasmodic index of preschool children was always lower than 50% in spite of an inoculation rate of about one infected bite per child per night all along the year. The actual incidence rate estimated with Muench model was ĥ = 0,015 while the recovery rate was r = 0,032 (i.e. about three times faster than the usual values admitted since Mac Donald work) for young children (0 to 4 years old). A computer study has shown that an incidence of h1 = 0,0012 was enough for the "infection" of children (Ross model) while an incidence of h2 = 0,0014 would induce a situation of superinfection (Dietz et al. model). Therefore the actual incidence was 10 to 12 times higher than the critical values of the incidence rate. To decrease the malaria transmission at a level lower than the critical values i.e. to obtain a reproduction rate below I the calculations and graphs have shown that anopheline density or human gametocytaemia have to be reduced by about 90% while the survival rate of the vectors must be reduced by about 12%. Therefore it appeared that the determination of the critical levels of every parameters of malaria transmission is a needful stage for a better planification of any malaria control programme.