It is essential for all practitioners to be aware of the antibiograms for pathogens in their communities/hospitals. It is rare that a pathogen is identified at the time of hospital admission. Accordingly, clinicians must make a best guess of the likely pathogens, based on the history, site of likely infection(s) and the host (normal or immunocompromised). As a general rule, empiric infections occurring in patients who have been hospitalized, in a nursing home or in and out of dialysis centers in the past 3-6 months, require treatment for staphylococcus and virulent gram negative rods like pseudomonas. Although there are no data to justify the process, when gram positive cocci are in the gram stain of a body fluid, we isolate the patient until methicillin resistant staphylococcus an vancomycin resistant enterococcus have been ruled out. We have a low-threshold for treating patients who are very ill, who have significant risks for MRSA (e.g. indwelling catheters, dialysis patients, nursing home patients) and with gram positive cocci in clusters with vancomycin until final identification and sensitivities of the pathogen are known.

The ENH practice a "rotating crop" strategy of antibiotic selection. Every 3-6 months, a committee examines the resistance patterns of various pathogens and we choose a "go to" empiric regimen that can be used when the suspicion of nosocomial pathogens is high. This technique may be useful to reduce the likelihood of multi-drug resistant pathogens (especially pseudomonas).

TO AVOID DRUG RESISTANCE, CLINICIANS SHOULD CHOOSE AN ANTIBIOTIC OF MORE NARROW SPECTRUM THAT KILLS THE IDENTIFIED PATHOGEN WHEN RESULTS OF CULTURES/SENSITIVITIES ARE KNOWN. ALSO, PATIENTS SHOULD BE TREATED FOR A COMPLETE COURSE AND ATTENTION SHOULD BE PAID THAT APPROPRIATE DOSES OF ANTIBIOTIC ARE USED AT THE MOST APPROPRIATE FREQUENCY TO MAXIMIZE KILLING AND PREVENT EMERGENT RESISTANCE.