Overall Study Question
The objective of the study was to determine if a targeted one-on-one educational program (“academic detailing”) by clinical specialists on a patient-specific basis could reduce excessive use of broad-spectrum antibiotics.
The study population consisted of the interns and residents who wrote antibiotic prescriptions for either levofloxacin or ceftazidime on 17 internal medicine services (9 general medicine, 4 oncology, and 4 cardiology) during a 18 week period. The educational intervention was directed at the interns and residents on the services in the intervention group who wrote “unnecessary” prescriptions for either levofloxacin or ceftazidime. A prescription was considered unnecessary if it fell outside the guidelines for appropriate use developed by the hospital’s Division of Infectious Diseases. One of the 3 clinical educators (2 Infectious Disease physicians and 1 clinical pharmacist) contacted the prescribing intern or resident to arrange a face-to-face meeting. Using an academic detailing approach, the clinical educator would present information in a case-relevant, concise manner, stressing microbiological data, local resistance patterns, and the clinical literature. The educator would suggest alternative regimens, but the final choice was left to the intern or resident. The educational sessions lasted approximately 10 minutes.
The primary outcome variable was to compare the average number of days of unnecessary levofloxacin and ceftazidime use during each 2-week period for the intervention and control services. Secondary outcome variables included examining the medication prescriptions for the day of the educational intervention and the subsequent day to determine if the unnecessary antibiotic was continued, discontinued or changed in route or dosage. Length of stay, proportion of patients transferred to ICU, readmitted within 30 days, or died in hospital were also compared for every patient admitted to an intervention or control service during the study period.
Are the Results of the Study Valid?
1. Was assignment of patients randomized?
Seventeen services were assigned to intervention or control group status using a blocked randomization design to achieve balanced patient groups.
Five general medicine, 2 oncology, and 2 cardiology services were included
in the intervention group. Four general medicine, 2 oncology, and 2 cardiology services
made up the control group.
2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?
During the 18-week study period, 490 patients from all study services were prescribed levofloxacin or ceftazidime. Computerized records indicated that 212 patients did not meet the criteria for ‘unnecessary’ use and were excluded. The remaining 278 prescriptions in 260 patients were flagged as unnecessary (125 in the intervention services and 153 in the control services). In 14 instances, the clinical educator decided not to contact the prescribing intern or resident after reviewing the patient’s chart. This was typically due to the fact that the previous Infectious Disease consultant had recommended one of the targeted antibiotics. These prescriptions were included in all analyses.
- 3. Were patients, their clinicians, and study personnel ‘blind’ to treatment?
- No. The interns and residents were not aware that their prescribing practices were being studied. While the clinical educators were unblinded, the initial screening criteria to identify unnecessary prescriptions were applied without regard to study group assignment.
- 4. Were the groups similar at the start of the trial?
- Patient characteristics (admissions per 2 week period, patient age and gender) were similar on the intervention and control services and did not differ during baseline and study periods. There were slightly higher rates of fever and neutropenia among patients prescribed levofloxacin on the intervention services during the study period.
- 5. Aside from the experimental intervention, were the groups treated equally?
- The intervention and control services were treated equally. The study hospital’s Division of Infectious Diseases developed a set of guidelines for first line antibiotic therapy including recommendations for the proper use of levofloxacin and ceftazidime. All house staff received a pocket sized laminated brochure of the guidelines prior to the study. The possibility of contamination between intervention and control groups exists as residents switch services every month. This would have been expected to bias the results against finding a difference or towards finding less of an effect.
What were the Results?
1. How large was the treatment effect?
During the baseline period, the mean number of days of unnecessary levofloxacin or ceftazidime use in a 2-week period was not different between the intervention and control services (8.5 +/- 7.8 days versus 7.6 +/- 4.7 days). During the study period, the intervention services had 5.5 +/- 2.1 days of unnecessary antibiotics versus 8.8 +/- 2.2 days on the control services (a relative reduction of 37%, p<0.001) as a result of 28 fewer starts, and a 0.8 day shorter average antibiotic course.
On the intervention services, 70% of the unnecessary prescriptions were discontinued the next day versus only 30% on the control services. Also, 55% of the patients on the intervention services had all of their antibiotics discontinued versus only 16% on control services. Route changes were equal and relatively uncommon on both services (14%).
2. How precise was the estimate of the treatment effect?
A multivariate analysis done to account for repeated measures of target antibiotics and baseline prescribing indicated that the risk of receiving a day of unnecessary levofloxacin or ceftazidime was reduced by 41% (CI: 44-78%) on the intervention services compared with control.
Will the Results Help Me in Caring for My Patients?
- 1. Can the results be applied to my patient care?
The relative success of ‘academic detailing’ is dependent upon the ability and the availability of the clinical educator. This study used 2 Infectious Disease physicians and 1 clinical pharmacist as their clinical educators. While many hospitals do not have an Infectious Disease Consult Service, or have physicians available to perform academic detailing, most hospitals have clinical pharmacists who routinely review the use of broad spectrum antibiotics on a patient-specific basis and consult with the prescribing physician when an alternative antibiotic may be more appropriate. While this paper does not differentiate between the effectiveness of the clinical educators, the authors mention that the impact of the clinical pharmacist was similar to that of the Infectious Disease physicians.
2. Were all clinically important outcomes considered?
All clinically important outcomes were considered.
- 3. Are the likely treatment benefits worth the potential harms and costs?
- The benefit of the academic detailing sessions was a decrease in the use of the targeted broad-spectrum antibiotics, levofloxacin and ceftazidime, with no measurable harmful effect. There was no difference in length of stay, proportion of patients transferred to ICU, readmitted within 30 days, or death rates between the 2 services. In addition, decreased use of these broad-spectrum agents should reduce the risk of antibiotic resistance associated with overuse of broad-spectrum antibiotics.
This is the first study to formally evaluate the effectiveness of academic detailing in a hospital setting. Previous studies involving academic detailing have primarily focused on altering prescribing practices in the community or residential care settings. This study reinforces the value of using a brief, face-to-face interaction with evidence-based information as an effective method for modifying prescribing practices.