A novel method of estimating cost of therapy by using patient population characteristics: analysis of fluoroquinolones in various populations with different distributions of renal function.

INTRODUCTION

Formulary decisions regarding a given drug class are often made in the absence of patient outcome and/or sophisticated pharmacoeconomic data. Analyses that consider factors beyond simple acquisition costs may be useful in such situations. For example, the cost implications of using manufacturers' recommendations for dosing in patients with renal dysfunction may be important, depending on the distribution of various levels of renal function within a patient population.

METHODS

Using four 1000-patient populations representing different renal function distributions and a fifth population of our medical center's distribution, we determined the costs of therapy for intravenous and oral levofloxacin, gatifloxacin, and moxifloxacin for a 10-day course of therapy for community-acquired pneumonia. Costs considered were average wholesale prices (AWPs), 50% of AWP, or same daily price, plus intravenous dose preparation and administration costs when applicable. Costs for each renal function distribution were examined for significant differences with an analysis-of-variance test. Also, costs of failing to adjust dosing regimens for decreased renal function were determined.

RESULTS

Differences in fluoroquinolone costs (AWP, 50% AWP, or when matched as the same daily price) among the populations were found. When considering same daily prices, differences among populations ranged from about 35,000 dollars with intravenous gatifloxacin to more than 51,000 dollars for intravenous levofloxacin (all fluoroquinolones, p>0.05). Within a population, differences in costs among the intravenous fluoroquinolones ranged from 47,000-99,000 dollars. Rank orders of the drugs and population costs of therapy were affected by the pricing structure used and varied by the specific population and drug. Differences among the fluoroquinolones or populations were much smaller (<2100 dollars) when considering oral regimens. Costs potentially incurred by failing to adjust dosing for renal function were substantial.

CONCLUSION

Formulary decisions can be facilitated by considering factors such as patient characteristics and related dosing in addition to simple acquisition costs. In our example, consideration of the distribution of renal function within a given patient population and related dosing for these fluoroquinolones revealed potentially important differences within the class.