Pharmaceutical Research

Studies examine the practice of prescribing medications to outpatients that can dangerously interact with one another

Many patient problems caused by medications, or adverse drug events, are due to dangerous interactions of coprescribed medications. A new study reveals that one-third of primary care patients are prescribed drugs that strengthen the blood-thinning effect of the anticoagulant warfarin. This increases a patient's risk of internal bleeding; however, using computerized drug interaction alerts can reduce such dangerous prescribing. A second study shows that between 18 and 28 percent of primary care patients are prescribed a drug that can adversely interact with warfarin or three other commonly used drugs. Both studies involved the HMO Research Network's Center for Education and Research in Therapeutics (CERTs), which is supported by the Agency for Healthcare Research and Quality (HS11843), and are summarized here.

Feldstein, A.C., Smith, D.H., Perrin, N., and others (2006, May). "Reducing warfarin medication interactions." Archives of Internal Medicine 166, pp. 1009-1015.

This study found that nearly a third of outpatients taking the anticoagulant warfarin were also prescribed another drug that dangerously increased its blood-thinning effect. Yet, when primary care doctors received computer alerts to such drug-drug interactions at the time of prescribing, the warfarin-interacting medication prescription rate was reduced by 15 percent.

The study included 239 primary care providers at 15 primary care clinics and 9,910 patients taking warfarin. All clinics received electronic medical record alerts for the coprescription of warfarin and one of five interacting medications (acetaminophen, nonsteroidal anti-inflammatory medications, fluconazole, metronidazole, and sulfamethoxazole). Seven clinics also received group academic detailing (a 40-minute educational session to a small group of clinicians about the clinical risks of coprescribing drugs that interact with warfarin, computerized alerts, and tools for later reference).

Coinciding with the alerts, there was an immediate and continued reduction in the warfarin-interacting medication prescription rate from 3,294 to 2,804 per 10,000 warfarin users per month. This was an overall decline of 15 percent over a 12-month period. Group academic detailing did not enhance the effectiveness of the computerized alerts. However, well-constructed alerts may be self-explanatory. They may provide the just-in-time training thought to be key to improving prescribing practice, explain the researchers.

Lafata, J.E., Schultz, L., Simpkins, J., and others (2006, June). "Potential drug-drug interactions in the outpatient setting." Medical Care 44(6), pp. 534-541.

This study found that from 7 to 28 percent of outpatients who were prescribed one of four drugs (warfarin, digoxin, cyclosporine, or lovastatin/simvastatin) were also prescribed another drug that put them at risk for a potentially dangerous drug-drug interaction. When extrapolated to the U.S. insured adult population, an estimated 1.3 to 2.7 million adults are dispensed a potentially interacting pair of medications. For example, prescribing nonsteroidal anti-inflammatory medications or amiodarone with warfarin can increase the risk of internal bleeding. The risk of an irregular heart beat (cardiac arrhythmia) is heightened when amiodarone or erythromycin are coprescribed with digoxin. Kidney toxicity can result when erythromycin or rifampin are prescribed with cyclosporine. Myopathy (progressive muscle weakness) can strike when cyclosporine or erythromycin are added to lovastatin or simvastatin.

In this study, 7 to 18 percent of primary care patients were coprescribed interacting drugs on the same day, and 18 to 28 percent were coprescribed interacting drugs during the "days supply," that is, the time period for which the patient had the other medication available. The findings were based on an examination of coprescribing over a 1-year period for 67,820 insured adults who were prescribed 1 of the 4 drugs studied and receiving care from one of the primary care clinics that were part of the HMO CERTs. The researchers used sample-based estimates to project the rate of dangerous coprescribing involving the four medications at a national level.

Return to Contents
Proceed to Next Article