The proper dosage schedule of antibiotics has generally been determined empirically, due to the difficulty of clinical trials. Initially, the dosage was chosen to allow high sustained levels greater than MIC in the blood. Antibiotics (beta lactams, tetracyclins, macrolides) were given at high doses three to six times daily, whatever their kinetic properties. The data obtained by Eagle3 with beta lactams in animal models of streptococcal and treponemal infections outlined the importance of interval between doses on the in vivo efficacy. They also showed that increasing the dose of penicillin had a positive effect on the bactericidal activity only through the persistence of effective levels (greater than MIC) at the site of infection. Further illustrations were given through experimental and clinical studies with beta lactams or other compounds on different types of infections: LRTIs, UTIs, meningitis, and endocarditis. The importance of both dynamic (i.e., pattern of bactericidal effect) and kinetic (elimination half-life) parameters was thus further identified. Information on toxicity with some compounds with a narrow therapeutic index, such as aminoglycosides, indicated that increasing the dose to enhance efficacy had some limitations. This led to numerous studies on the relations between concentration and toxicity, stating that nephro- or ototoxicity were not directly related to peak level in serum. Experimental studies showed that OD administration of aminoglycosides was both more efficient and less toxic than the multiple-dose regimen of the same daily amount. Economic considerations progressively justified attempts to both reduce the dose and the work load related to multiple administrations.(ABSTRACT TRUNCATED AT 250 WORDS)