Early benefit assessment (EBA) in Germany: analysing decisions 18 months after introducing the new AMNOG legislation

Abstract

Objectives: Since the introduction of the German health care reform in January 2011, an early benefit assessment (EBA) is required for all new medicines. Pharmaceutical manufacturers have to submit a benefit dossier for evaluation by the Institute for Quality and Efficiency in Health Care (IQWiG). A final decision is made by the Federal Joint Committee (G-BA). The aim of this investigation was to analyse the outcomes 18 months after introduction of the new legislation and to identify critical areas requiring further discussion and development.

Methods: All EBAs commenced prior to June 2012 were included. The G-BA website was used to obtain manufacturers' benefit dossiers, IQWiG assessments, and G-BA decisions. Four areas of interest were analysed: levels of additional benefit, appropriate comparative therapy (ACT), patient-relevant endpoints, and adverse events.

Results: Twenty-seven EBAs were analysed. IQWiG stated a benefit in 50% of EBAs, whereas G-BA stated a benefit in 63%, but only in 50% of identified subgroups and 40% of patients involved. In 12 EBAs, the ACT suggested by G-BA differed from the comparator used in phase III trials. The G-BA reported no benefits on health-related quality of life. Discrepancies arose in morbidity outcomes such as 'progression-free survival' and 'sustained virological response'. Categorisation and balancing of adverse events was conducted within various assessments.

Conclusions: Considerable variance was observed in the levels of additional benefit reported by pharmaceutical manufacturers, IQWiG and G-BA. The areas of disagreement included ACT selection, definition of subgroups and patient-relevant endpoints, and classification and balancing of adverse events.

Fig. 1

Flow chart covering benefit assessment and price negotiation according to the new German regulations since January 2011

Fig. 2

Presence of additional benefit as reported by IQWiG, G-BA and HAS according to the number of products evaluated (n)

Fig. 3

Acceptance of ACT selected by G-BA and consequences on the added benefit decision

Fig. 4

Evaluation of negative patient-relevant outcomes by G-BA in the included EBAs. For products at the end of the arrows this evaluation leads to an up- and downgrade, respectively. This evaluation does not influence the benefit level of the products before the gap

Fig. 5

Level of additional benefit (%) in 40 assessed patient subgroups in the 23 evaluated EBAs (23 of 31 drugs, 4 drugs exempted and 4 without submitted dossier)

Fig. 6

Level of additional benefit compared to % and total number of patients in 39 (Rilpivirine and Emtricitabine/Rilpivirine/Tenofovir disoproxil are referring to the same population) assessed patient subgroups in the 23 (23 from 31 drugs, 4 exempted drugs and 4 drugs without submitted dossier) evaluated EBAs. Total (sum of mean values from G-BA decisions: 1,408,742 patients) (Without the Linagliptin population of 1,219,500 patients due to the dossier re-submission)