APPLYING THE EFFICIENCY FRONTIER IN PHARMACEUTICAL POLICY MAKING. A CASE STUDY IN TREATMENT RESISTANT HIV/AIDS PATIENTS IN THE GERMAN SETTING
Author(s)
Lieven Annemans, Professor, Professor of Health Economics1, Andrew Hill, PhD, Managing Director2, Erik Smets, MSc, MBA, Director Health Economnics, Outcomes Research & Pricing3, Silas C Martin, MS, Director31University of Ghent, Brussels University, Ghent, Belgium; 2 HIV Solutions, London, United Kingdom; 3 Johnson & Johnson Pharmaceutical Services LLC, Mechelen, Belgium
OBJECTIVES Recently, the German IQWIG has proposed the efficiency frontier approach to support decisions regarding reimbursed price levels, and published recommended methods in 2008. Treatment regimens appearing on the efficiency frontier are considered the most economically efficient. Our aim was to apply these methods in an economic evaluation of pharmaceutical options in treatment-experienced HIV/AIDS patients. METHODS We identified 10 published randomized controlled trials for the target population (POWER 1&2, RESIST 1&2, MOTIVATE 1&2, DUET 1&2, BENCHMRK 1&2), of which the following information was extracted for all arms: baseline characteristics; % of patients with a viral load <50 copies/mL at week 48 (response rates); enfuvirtide (ENF) use as co-medication and its impact on response; and the use of antiretroviral therapies during the studies. Unit drug costs were obtained from Rote Liste©. A one year time horizon was chosen. The results of all treatment arms (average and + or – ENF) were plotted on a coordinate system with annual drug costs per patient on the horizontal axis and response rates on the vertical axis, the latter after adjustment for baseline characteristics using logistic regression on the pooled raw data of the DUET trials. RESULTS Twenty-six cost-efficacy points were created representing all compared strategies. Drug costs per year per patient varied between €22,186 and €61,715. Response rates varied between 8.4% and 68.7%. All interventions appearing on the frontier were combinations including etravirine. The last line segment of the frontier had a slope of €1354 per extra percentage response. CONCLUSIONS Constructing an efficiency frontier plot was feasible in this indication. However, detailed baseline characteristics and logistic regression were required for adjustment. Regimens containing etravirine appear to be economically efficient. Longer term evaluation including all health care costs could add valuable information, but would require many assumptions given the limited available data for the 26 compared strategies.
Conference/Value in Health Info
2009-05, ISPOR 2009, Orlando, FL, USA
Value in Health, Vol. 12, No. 3 (May 2009)
Code
PIN23
Topic
Economic Evaluation
Topic Subcategory
Cost-comparison, Effectiveness, Utility, Benefit Analysis
Disease
Infectious Disease (non-vaccine)