In a previous study, we found that unsolicited recommendations to physicians of medically stable patients with pneumonia to suspend parenteral antimicrobials shortened hospital length of stay (LOS) significantly. In this study, we made similar recommendations to physicians treating patients with different indications for parenteral antimicrobials, to examine the effect on LOS.
A nurse-interventionist presented randomly assigned physicians with nonconfrontational suggestions to discontinue parenteral antimicrobials by substituting comparable oral antimicrobials or stopping treatment. Patients were being treated for urinary tract infection, skin infection, or no evident infection. Blinded observers evaluated in-hospital and 30-day postdischarge patient courses. Methodologies were identical to the previous study.
There were 70 physician-patient episodes (49 intervened episodes, 21 control episodes). In 44 episodes (90%), compliant physicians discontinued parenteral antimicrobials. Compared to a median postrandomization LOS of 2.5 days (range, 0 to 40.5) for 21 patients of control physicians, the corresponding LOS for 44 patients of compliant physicians was two days (range, 0 to 8; P= 1.0), and for five patients of noncompliant physicians, five days (range, 3 to 11; P= 0.04).
The combined occurrence of all adverse events detected in this and the previous study was 11% for patients of control physicians, compared to 14% for patients of compliant physicians (P= 0.2), and 19% for patients of noncompliant physicians (P<0.05).
For patients of compliant physicians hospitalized with urinary tract infection, skin and soft tissue infection, or no evident infection, cessation of parenteral antimicrobials did not significantly shorten LOS, due to brief LOS of patients of control physicians. Patients of noncompliant physicians experienced more adverse events and prolonged LOS. The appropriateness of routine continuous use of parenteral antimicrobials in medically stable inpatients is questioned.