OBJECTIVES: Since the introduction of AMNOG more than 100 IQWiG recommendations and G-BA decisions on the early benefit assessment have been published. In various assessments, the IQWiG recommendation and the related G-BA decision differ in terms of argumentation and result. In this study, these differences were systematically analyzed in order to draw conclusions for further proceedings.  METHODS: Data of all published IQWiG recommendations and G-BA decisions were standardized and transferred into the Prismaccess database. In a systematic analysis, major differences between the recommendations and decisions were quantitatively assessed. All documents of the identified decisions where differences were observed were then compared qualitatively.  RESULTS: In several aspects, major differences were found between IQWiG recommendations and G-BA decisions in approximately 25% of all cases. On the subgroup level, it was observed in at least n=10 cases that the G-BA tends to use a broader definition of subgroups in its decisions than used in the IQWiG dossiers, (e.g. Telaprevir). In terms of the extent of the additional benefit, 17 differences between IQWiG and G-BA decisions were registered, while mostly G-BA tend to evaluate better than IQWiG. In the probability of the additional benefit also some differences were observed. This can be explained by the opportunity to provide additional/new evidence during the oral hearing, which may influence the decision positively. Some proceedings were identified that took another way, as intended by the legislature: for example with Cobicistat, an incomplete paperwork was handed in, to achieve at least the price of the ACT. Still accepted by the G-BA, it is obvious that this should not be the normal case.  CONCLUSIONS: In an early stage, G-BA consultations might be helpful clarifying subgroups-requirements. In the process, the oral hearing was identified as important instrument to supplement the required data and to correct uncertain issues from the dossier.