How pharmacists respond to on-line, real-time DUR alerts.

OBJECTIVE: To quantify the type and frequency of drug utilization review (DUR) alerts sent by one claims processor to pharmacists; identify how pharmacists respond to these on-line, real-time DUR messages; and quantify the interventions taken by these pharmacists as a result of these alerts. DESIGN: Retrospective analysis of Medicaid claims from July 1, 1995, through June 30, 1996. SETTING: State of Delaware. PARTICIPANTS: 55,000 Medicaid recipients served by 170 participating pharmacies and 2,000 physicians. INTERVENTIONS: All on-line DUR alerts sent to pharmacists and the pharmacists' responses were categorized by alert type and analyzed by drug class. MAIN OUTCOME MEASURES: Pharmacists' response (dispensed prescription, contacted prescriber, talked with patient, consulted own reference sources) and drug classes. RESULTS: During the study period, 807,017 claims generated 83,260 DUR alerts involving 73,554 (9.1%) prescriptions. Prescriptions were not dispensed in 20.9% of cases because of the DUR message. Prescriptions were dispensed 17.7% of the time after the pharmacist contacted the prescriber, in 20.6% of cases after the pharmacist talked with the patient, and 37.2% of the time after reviewing internal resources. Action taken by pharmacists varied among and within DUR criteria categories. Specific examples of alerts generated in high-frequency and high-profile areas are reviewed, some of which generated inconsistent responses. CONCLUSION: Among Delaware Medicaid providers, drug alerts resulted in pharmacists not dispensing prescriptions in a surprising percentage of situations. Pharmacists' responses varied according to the drug class involved and the type of alert received.