27PEC ECONOMIC EVALUATION STUDIES SUBMITTED TO INFORM CANCER DRUG FUNDING DECISIONS: A CLOSER LOOK AT THE QUALITY AND LIMITATIONS

Wednesday, October 22, 2008
Columbus A-C (Hyatt Regency Penns Landing)
Jean Hai Ein Yong, MASc1, Sandjar Djalalov, PhD1 and Jeffrey Hoch, PhD2, (1)Cancer Care Ontario, Toronto, ON, Canada, (2)St. Michael's Hospital, University of Toronto, Toronto, ON, Canada
Purpose: To understand the limitations of economic evaluation studies submitted to the interim Joint Oncology Drug Review (JODR) to inform oncology drug reimbursement decisions in Canada. Methods: We reviewed the pharmacoeconomic reviewers' comments on all submissions to Canada's JODR from March 2007 to March 2008.  We extracted comments related to limitations of the analysis into data abstraction templates and analyzed them. Results: Reviewers' comments on 15 JODR submissions were examined. The most commonly mentioned limitation was insufficient sensitivity analysis to inform decisions (85%). This includes missing sensitivity analysis around key parameters or the range used in the sensitivity analysis was not justified. Seventy-percent of the submissions had calculation errors, invalid representation of disease or did not incorporate data correctly in the model. Other important limitations include issues around inappropriate comparator or outcome, costing methodology and assumptions, quality of clinical evidence and health state utility estimates. Conclusion: Quality of economic evaluation is often cited as a common barrier to the use of economic evaluation in drug policy decision making. This study allows us to better understand the limitations in the economic evaluation studies submitted to inform oncology drug reimbursement decisions in Canada. This study is the first step to improving the quality and relevance of economic evaluation to better inform drug reimbursement decisions. A potential use of our findings
will be to inform the research in developing new ways of communicating methodology guidelines for economic evaluation.