Multiresistant Gram-negative bacilli, including strains of Klebsiella pneumoniae, Enterobacter spp, Acinetobacter baumannii and Pseudomonas aeruginosa, resistant to broad spectrum beta-lactams, aminoglycosides and fluoroquinolones, are recovered at increasing frequency from patients suffering from nosocomial infections, particularly from those receiving intensive care. The emergence and spread of resistant pathogens to endemic and epidemic levels has frequently been related in time and place to the intensive use of antibiotics to which these microorganisms have developed resistance, notably third generation cephalosporins and fluoroquinolones. Recent investigations have indicated that the prevalence of resistance can be reduced by scheduled changes of empiric treatment regimens, involving discontinuation of intensively prescribed drugs and substitution with newly introduced antibiotics of another class to which the prevalent resistant strains remain susceptible. Among these drugs, penicillins---beta-lactamase inhibitor combinations, 'fourth generation' cephalosporins and, where little used previously, fluoroquinolones, have been introduced successfully in high risk units where ceftazidime-resistant strains of K.pneumoniae, Enterobacter and Citrobacter spp or glycopeptide-resistant enterococci had become highly prevalent. However, these studies do not demonstrate a direct causal relationship between changes in prescribing practices and ecological improvements, because their observational design cannot be controlled. In most studies, several important factors influencing the dynamics of resistance were not monitored and the relative contribution of decreased emergence versus control of cross-transmission to the improved susceptibility rates is not clear. We propose that additional long-term studies are required to better track the ecological impact and to determine the optimal modalities of programmed changes of antibiotic prescribing as an antibiotic resistance prevention or control strategy.