A computer-assisted management program for antibiotics and other antiinfective agents

• HIT Description: A computerized health record that included an anti-infective management program that alerts physicians to the latest pertinent information on the individual patient at the time therapeutic decisions are made More info...
  
• Purpose of Study: Improve the use of antibiotics and other anti-infective agents
  
• Years of study: 1992-1995
  
• Study Design: Cohort
  
• Outcomes: Utilization of services, costs of care
  

• Settings: The 12-bed intensive care unit (ICU) of an academically affiliated 520-bed acute care hospital.
• Intervention: The authors compared utilization and costs of care of patients admitted to the ICU during a period of time before and after the implementation of the anti-infective management program. The program addressed numerous patient-specific and disease specific issues, including costs of anti-infective agents, alerts and suggestions regarding the dose, route and duration of therapy, drug-drug interactions, drug-laboratory test interactions, drug-nutrient interactions, drug-therapy omissions, and indications for drug use.
• Healthcare Utilization: During the one-year intervention period, physicians in the ICU used the management program on average 14.3 times per day, and the program was used to order anti-infective agent regiments 942 times. Physicians prescribed the computer suggested agents 46% of the time and they followed the computer suggested dosing and interval for 93% of the orders. Compared to the pre-intervention control period there was a slight increase in the overall use of antibiotics (73% during the intervention period, 67% during the control). During the intervention period, susceptibility-mismatch problems dropped from 206 episodes in the pre-intervention control period to 12 in the intervention period (p 0.01). Allergy alerts were generated 35 times in the intervention period compared with 146 in the control period, and alerts of excessive dosage of anti-infective agents were generated 87 times in the intervention period as compared with 405 times in the control period. In addition, during the pre-intervention control period the