The prevalence of antimicrobial resistance may vary dramatically between countries and regions and even among different patient populations and between hospitals in the same community. In light of these variations, physicians in practice must make a clinical judgment about the likely pathogen(s) involved in the infectious process. They should have access to up-to-date data on the antimicrobial resistance pattern of commonly encountered pathogens. Such data should be subdivided and analyzed for isolates from community as opposed to hospitalized patients. The data on intensive care units patients and from special groups of patients, such as those undergoing dialysis, organ transplant recipients, or cancer patients receiving chemotherapy, should also be made available to practicing physicians. When antimicrobial resistance is considered to be probable or highly likely, or if empirical therapy has failed, a concerted effort should be made to make a microbiologic diagnosis which will ultimately reduce the cost, simplify therapy and improve the outcome over treatment that is solely empirically based.