Utilization and Evaluation of Delphi Panels In German Amnog Assessments

Abstract

Expert advice and/or opinion can be a valuable source of information, especially if reliable evidence is not available. A Delphi Panel (DelP) is a structured approach, utilizing the expertise of experts to provide evidence in a consensus-driven approach. In accordance with the rules of procedure by the German Federal Joint Committee (G-BA) during the Benefit Assessment (AMNOG), expert advice achieved by consensus, in principle, is accepted within value dossiers. Aim of this study was to examine how DelPs are integrated in German AMNOG assessments and how they are evaluated by the Institute for Quality and Efficiency in Health Care (IQWiG) and the G-BA. Available AMNOG dossiers up until June 2017 were evaluated with respect to the integration of DelPs, size of the panel, acceptance and evaluation by IQWIG and G-BA, as well as decision on added benefit. Descriptive statistics were applied and the assessment of respective DelPs by IQWiG was analyzed. In total, n=251 AMNOG dossiers were included. Overall, DelPs were utilized in 4.4% (n=11) of all dossiers. Results of DelPs were incorporated in Module 3 (n=8) (epidemiology, cost, disease burden) and Module 4 (n=3) (clinical data). DelPs were used to address evidence needs in oncologic disorders (n=5), metabolic diseases (n=3), mental disorders (n=2), and diseases of the musculoskeletal system (n=1). On average 10.8 experts were included in the exercise. IQWIG, in the majority of cases, was basically positive, criticism focused on number of experts in two dossiers, as well as lacking discussion of limitations and transfer of results to target populations. In four dossiers IQWIG was not able to validate the indicated findings. DelPs are rarely used within the German AMNOG process. As the acceptance seems reasonable, the application of DelPs offers a viable option to address or minimize evidence gaps in an AMNOG assessment.