The evaluation of clinical practice includes three successive phases: demonstration of suboptimal clinical practice, proposals for improvement, and assessment of effective changes. We present a medical computerized database used for this evaluation.
Patient and methods: Our database includes 24 parameters for all hospitalized patients among which diagnosis and antibiotherapy. The first assessment phase indicated an unjustified heterogeneous antibiotherapy for community-acquired pneumonia (CAP). The second phase was the drafting of our own consensus for CAP treatment, aiming at reducing the use of fluoroquinolones and third generation cephalosporin. The computerized database was used in the third phase to check observance of our consensus.
Results: Two hundred and fifty patients were hospitalized for CAP before the consensus, from October to January 2005, 2006, and 2007, compared to 113 patients from October 2008 to March 2009. The rate of adequate prescriptions improved from the first period to the second one: 58% versus 69%, p=0.045. The inadequate use of fluoroquinolones and third generation cephalosporins was more frequent during the first period, compared to the second one: 49% versus 38%, p=0.048. The mean hospital stay was longer before applying our consensus: 9+/-5 days versus 7+/-4 days, p=0.009. Evolution was unfavorable in 15/250 cases (6%) and 3/113 cases (<1%), respectively.
Conclusion: A medical database allows for a rapid implementation of two out of three phases of clinical evaluation, through the appropriation of a new consensus, with a reduction of antibiotic misuse without impact on infectious morbidity and mortality.
(c) 2009 Elsevier Masson SAS. All rights reserved.